Documentos de Académico
Documentos de Profesional
Documentos de Cultura
January 1999
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
OFFICE OF
WATER
MEMORANDUM
I am pleased to provide you with the Environmental Protection Agency’s (EPA) guidance
document on the monitoring and modeling of combined sewer overflows (CSO) and their
impacts on receiving waters. This is the seventh in a series of guidance manuals that EPA
prepared to support implementation of the 1994 Combined Sewer Overflow Control Policy.
This manual presents a set of guidelines that provide flexibility for a municipality to
develop a site-specific strategy for characterizing its combined sewer system operations and
impacts and for developing and implementing a long-term CSO control plan. It is not a “how-
to” manual defining how many samples to collect or which flow metering technologies to use.
EPA used a peer-review process and solicited comments from CSO stakeholders and the
general public. The EPA identified the Water Environment Research Foundation (WERF) and
two technical experts to provide technical and scientific peer review. WERF convened a panel of
its technical experts to review the document. The peer reviewers and the other reviewers
submitted detailed comments and recommendations. EPA will make available to interested
parties a “response-to-comments” document detailing how it addressed comments received
during the peer review and the public comment period. I am very grateful to the peer reviewers
and the other individuals and organizations who participated in preparation and review. I believe
that this manual will assist municipalities as they develop and implement long-term CSO control
plans to meet the requirements of the Clean Water Act and the objectives of the EPA’s CSO
Control Policy.
If you have any questions on the manual or its distribution, please contact Tim Dwyer in
the Office of Wastewater Management at (202) 260-6064. Mr. Dwyer’s e-mail address is
dwyer.tim@epa.gov.
Recycled/Recyclable l Printed with Vegetable Oil Based Inks on 100% Recycled Paper (40% Postconsumer)
ACKNOWLEDGMENTS
The U.S. Environmental Protection Agency (EPA) wants to thank the Cities of
Columbus, Georgia; South Bend, Indiana; and Indianapolis, Indiana for allowing EPA to use
their experiences in monitoring and modeling as case studies for this manual. The experiences of
these cities provide excellent examples of the monitoring and modeling process associated with
developing and implementing combined sewer overflow (CSO) control programs. EPA believes
that use of case studies greatly enhances the value of the document.
EPA also acknowledges the peer reviewers who kindly donated their time and knowledge
to improving the technical and scientific discussions in this manual. The peer reviewers were
David Dilks, Limno-Tech, Inc.; Raymond M. Wright, Ph.D., P.E., University of Rhode Island;
and John Marr, Limno-Tech, Inc., who reviewed it on behalf of the Water Environment Research
Foundation.
Finally, EPA thanks those individuals and organizations that took the time and energy to
review and submit comments as part of the public review process. They are to be commended
for their perseverance and dedication to a long and arduous task.
EPA believes that the peer review process and the public comments greatly improved the
technical and scientific aspects of the manual. We hope that users will find the information in
the manual useful as they develop and implement CSO control plans.
Assistance in developing this manual was provided to EPA under contract number
68-C4-0034.
NOTICE
i January 1999
TABLE OF CONTENTS
Page
1. INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1
ii January 1999
TABLE OF CONTENTS (Continued)
Page
6.1 THE CSO CONTROL POLICY AND RECEIVING WATER MONITORING . 6-1
6.2 RECEIVING WATER HYDRAULIC MONITORING . . . . . . . . . . . . . . . . . . . . 6-2
6.2.1 Hydraulic Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
6.2.2 Analysis of Hydraulic Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6
6.3 RECEIVING WATER QUALITY MONITORING . . . . . . . . . . . . . . . . . . . . . . . 6-7
6.3.1 Water Quality Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-7
6.3.2 Analysis of Water Quality Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-9
6.4 RECEIVING WATER SEDIMENT AND BIOLOGICAL MONITORING . . . . 6-10
6.4.1 Sediment Sampling Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10
6.4.2 Analysis of Sediment Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.3 Biological Sampling Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-12
6.4.4 Analysis of Biological Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-16
iv January 1999
TABLE OF CONTENTS (Continued)
Page
8.1 THE CSO CONTROL POLICY AND RECEIVING WATER MODELING . . . . 8-1
8.2 MODEL SELECTION STRATEGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2
8.2.1 Hydrodynamic Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-3
8.2.2 Receiving Water Quality Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
8.3 AVAILABLE MODELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-5
8.3.1 Model Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8
8.3.2 Computer Models Supported by EPA or Other Government Agencies . 8-17
8.4 USING A RECEIVING WATER MODEL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-22
8.4.1 Developing the Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-22
8.4.2 Calibrating and Validating the Model . . . . . . . . . . . . . . . . . . . . . . . . . . 8-22
8.4.3 Performing the Modeling Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-23
8.4.4 Modeling Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-23
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . R-1
APPENDIX A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1
APPENDIX B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B- 1
V January 1999
TABLE OF CONTENTS (Continued)
Page
LIST OF EXHIBITS
Page
vi January 1999
LIST OF EXHIBITS (Continued)
Page
LIST OF EXAMPLES
INTRODUCTION
1.1 BACKGROUND
Combined sewer systems (CSSs) are designed to carry sanitary sewage (consisting of
domestic, commercial, and industrial wastewater) and storm water (surface drainage from rainfall
or snowmelt) in a single pipe to a treatment facility. CSSs serve about 43 million people in
approximately 950 communities nationwide, most of them located in the Northeast and Great Lakes
regions. During dry weather, CSSs convey domestic, commercial, and industrial wastewater to a
publicly owned treatment works (POTW). In periods of rainfall or snowmelt, total wastewater flows
can exceed the capacity of the CSS or the treatment facilities. When this occurs, the CSS is designed
to overflow directly to surface water bodies, such as lakes, rivers, estuaries, or coastal waters. These
overflows-called combined sewer overflows (CSOs)-can be a major source of water pollution.
Historically, the control of CSOs has proven to be extremely complex. This is partly due to
the difficulty in quantifying CSO impacts on receiving water quality and the site-specific variability
in the volume, frequency, and characteristics of CSOs. In addition, the financial considerations for
communities with CSOs can be significant. The U.S. Environmental Protection Agency (EPA)
estimates the CSO abatement costs for the 950 communities served by CSSs to be approximately
$45 billion based on results from the 1996 Clean Water Needs Survey.
To address these challenges, EPA issued a National Combined Sewer Overflow Control
Strategy on August 10, 1989 (54 Federal Register 37370). This Strategy reaffirmed that CSOs are
point source discharges subject to National Pollutant Discharge Elimination System (NPDES) permit
requirements and to Clean Water Act (CWA) requirements. The CSO Strategy recommended that
all CSOs be identified and categorized according to their status of compliance with these
requirements. It also set forth three objectives:
l Ensure that if CSOs occur, they are only as a result of wet weather
l Bring all wet weather CSO discharge points into compliance with the technology-based
and water quality-based requirements of the CWA
l Minimize the water quality, aquatic biota, and human health impacts of CSOs.
In addition, the CSO Strategy charged all States with developing state-wide permitting strategies
designed to reduce, eliminate, or control CSOs.
Although the CSO Strategy was successful in focusing attention, it failed to resolve many
fundamental issues. In mid-1991, EPA initiated a process to accelerate implementation of the
Strategy. The process included negotiations with representatives of the regulated community, State
regulatory agencies, and environmental groups. These negotiations were conducted through the
Office of Water Management Advisory Group. The initiative resulted in the development of a CSO
Control Policy, published in the Federal Register on April 19, 1994 (59 Federal Register 18688).
l Ensure coordination among the appropriate parties in planning, selecting, designing, and
implementing CSO management practices and controls to meet the requirements of the
CWA
The CSO Control Policy contains provisions for developing appropriate, site-specific NPDES
permit requirements for all CSSs that overflow due to wet weather events. It also announces an
enforcement initiative that requires the immediate elimination of overflows that occur during dry
weather and ensures that the remaining CWA requirements are complied with as soon as possible.
The CSO Control Policy contains four key principles to ensure that CSO controls are cost-
effective and meet the requirements of the CWA:
l Provide clear levels of control that would be presumed to meet appropriate health and
environmental objectives
l
Provide sufficient flexibility to municipalities, especially those that are financially
disadvantaged, to consider the site-specific nature of CSOs and to determine the most
cost-effective means of reducing pollutants and meeting CWA objectives and
requirements
l
Review and revise, as appropriate, WQS and their implementation procedures when
developing long-term CSO control plans to reflect the site-specific wet weather impacts
of CSOs.
In addition, the CSO Control Policy clearly defines expectations for permittees, State WQS
authorities, and NPDES permitting and enforcement authorities. These expectations include the
following:
l Permittees should immediately implement the nine minimum controls (NMC), which are
technology-based actions or measures designed to reduce CSOs and their effects on
receiving water quality, as soon as practicable but no later than January 1, 1997.
l Permittees should develop long-term control plans (LTCPs) for controlling CSOs. A
permittee may use one of two approaches: 1) demonstrate that its plan is adequate to
meet the water quality-based requirements of the CWA (“demonstration approach”), or
2) implement a minimum level of treatment (e.g., primary clarification of at least
85 percent of the collected combined sewage flows) that is presumed to meet the water
quality-based requirements of the CWA, unless data indicate otherwise (“presumption
approach’).
l WQS authorities should review and revise, as appropriate, State WQS during the CSO
long-term planning process.
Exhibit l-l illustrates the roles and responsibilities of permittees, NPDES permitting and
enforcement authorities, and State WQS authorities.
In addition to these key elements and expectations, the CSO Control Policy also addresses
important issues such as ongoing or completed CSO control projects, public participation, small
communities, and watershed planning.
Permittee NPDES Permitting Authority NPDES Enforcement Authority State WQS Authorities
l Evaluate and implement NMC l Reassess/revise CSO permitting l Ensure that CSO requirements and l Review WQS in CSO-impacted
strategy schedules for compliance are receiving water bodies
l Submit documentation of NMC incorporated into appropriate
implementation by January 1, 1997 l Incorporate into Phase I permits enforceable mechanisms l Coordinate review with LTCP
CSO-related conditions (e.g., NMC development
l Develop LTCP and submit for implementation and documentation l Monitor adherence to January 1,
review to NPDES permitting and LTCP development) 1997, deadline for NMC l Revise WQS as appropriate:
authority implementation and documentation
l Review documentation of NMC Development of site-specific
l Support the review of WQS in implementation l Take appropriate enforcement criteria
CSO-impacted receiving water action against dry weather
bodies l Coordinate review of LTCP overflows Modification of designated use to
components throughout the LTCP
l Comply with permit conditions development process and l Monitor compliance with Phase I, - Create partial use reflecting
based on narrative WQS accept/approve permittee’s LTCP Phase II, and post-Phase II permits specific situations
and take enforcement action as - Define use more explicitly
l Implement selected CSO controls l Coordinate the review and revision appropriate
from LTCP of WQS as appropriate Temporary variance from WQS
To help permittees and NPDES permitting and WQS authorities implement the provisions
of the CSO Control Policy, EPA has developed the following guidance documents:
Combined Sewer Overflows - Guidancefor Long-Term Control Plan (U.S. EPA, 1995a)
(EPA 832-B-95-002)
Combined Sewer Overflows - Guidance for Nine Minimum Controls (U.S. EPA, 1995b)
(EPA 832-B-95-003)
Combined Sewer Overflows - Guidance for Screening and Ranking (U.S. EPA, 1995c)
(EPA 832-B-95-004)
Combined Sewer Overflows - Guidance for Funding Options (U.S. EPA, 1995d) (EPA
832-B-95-007)
Combined Sewer Overflows - Guidance for Permit Writers (U.S. EPA, 1995e) (EPA
832-B-95-008)
EPA has printed a limited number of copies of each guidance document and has made them
available through several sources:
Electronic copies of some of the guidance documents are also available on EPA’s Office of
Water Internet site (http://www.epa.gov/ow/).
This manual explains the role of monitoring and modeling in the development and
implementation of a CSO control program. It expands discussions of monitoring and modeling
introduced in the CSO Control Policy and presents examples of data collection and CSS simulation.
This manual is not a “how-to” manual defining how many samples to collect or which flow
metering technologies to use. Rather, it is a set of guidelines that provides flexibility for a
municipality to develop a site-specific strategy for characterizing its CSS operation and impacts and
for developing and implementing a comprehensive CSO controlplan. CSSs vary greatly in their
size, structure, operation, and receiving water impacts. A monitoring and modeling strategy
appropriate for a large city such as New York or San Francisco would generally not apply to a small
CSS with only one or two flow regulators and outfalls. In addition, communities have varying
degrees of knowledge about how their CSSs react hydraulically to wet weather and how their CSOs
affect receiving water quality. A municipality that does not know the location of its CSO outfalls
has different information collection needs from a municipality that has already conducted CSS flow
and water quality studies.
This manual provides guidance for communities of all sizes. It presents low-cost monitoring
and modeling techniques, which should prove particularly helpful to small communities. However,
communities with large CSSs should note that inexpensive techniques often prove useful in
extending monitoring resources and in verifying the performance of more sophisticated techniques
and equipment.
To use this manual, a municipality should already be familiar with the basic functioning of
its CSS, basic monitoring procedures, and the general purpose of modeling. Since basic monitoring
and modeling techniques are already covered extensively in other technical literature, this manual
focuses mainly on the process of characterization as described in the CSO Control Policy, referring
to other literature for more in-depth explanations of specific techniques or procedures.
This manual begins with an overview of monitoring and modeling under the CSO Control
Policy, and then provides a detailed discussion of the monitoring and modeling activities that should
be conducted for NMC implementation and LTCP development and implementation. These
activities (and the chapters in which they are discussed) are as follows:
Monitoring and modeling activities are central to implementation of the CSO Control Policy.
Thoughtful development and implementation of a monitoring and modeling plan will support the
selection and implementation of cost-effective CSO controls and an assessment of their
improvements on receiving water quality.
This chapter describes general expectations for monitoring and modeling activities as part
of a permittee’s CSO control program. It also describes how monitoring and modeling efforts
conducted as part of CSO control program implementation can be coordinated with other key EPA
and State programs and efforts (e.g., watershed approach, other wet weather programs).
While this chapter will describe general expectations, EPA encourages the permittee to take
advantage of the flexibility in the CSO Control Policy by developing a monitoring and modeling
program that is cost-effective and tailored to local conditions, providing adequate but not duplicative
or unnecessary information.
The CSO Control Policy urges permittees to develop a thorough understanding of the
hydraulic responses of their combined sewer systems (CSSs) to wet weather events. Permittees may
also need to estimate pollutant loadings from CSOs and the fate of pollutants in receiving water both
for existing conditions and for various CSO control options. The CSO Control Policy states that
permittees should immediately undertake a process to characterize their CSSs, demonstrate
implementation of the nine minimum controls (NMC), and develop a long-term CSO control plan.
Characterizing the CSS and its hydraulic response to wet weather events, implementing the NMC
and producing related documentation, and developing a long-term control plan (LTCP) will involve
gathering and reviewing existing data, and, in most cases, conducting some field inspections,
monitoring, and modeling. Since flexibility is a key principle of the CSO Control Policy, these
activities will be carried out to different degrees based on each permittee’s situation. In particular,
the type and complexity of necessary modeling will vary from permittee to permittee.
The CSO Control Policy recommends that a Phase I permit require the permittee to
immediately implement technology-based requirements, which at a minimum include the NMC, as
determined on a best professional judgment (BPJ) basis by the NPDES permitting authority. The
NMC are:
1. Proper operation and regular maintenance programs for the sewer system
4. Maximization of flow to the publicly owned treatment works (POTW) for treatment
7. Pollution prevention
8. Public notification to ensure that the public receives adequate notification of CSO
occurrences and CSO impacts
9. Monitoring to effectively characterize CSO impacts and the efficacy of CSO controls.
The NMC are technology-based controls, applied on a site-specific basis, to reduce the
magnitude, frequency, and duration of CSOs and their impacts on receiving water bodies. NMC
measures typically do not require significant engineering studies or major construction and thus
implementation was expected by January 1, 1997. EPA’s guidance document Combined Sewer
Overflows - Guidance for Nine Minimum Controls (U.S. EPA, 1995b) provides a detailed
description of the NMC, including example control measures and their advantages and limitations.
l Mapping the drainage area for the CSS, including the locations of all CSO outfalls and
receiving waters
l Identifying, for each receiving water body, designated and existing uses, applicable water
quality criteria, and whether water quality standards (WQS) are currently being attained
for both wet weather and dry weather
l Compiling existing information on water quality impacts associated with CSOs (e.g.,
beach closings, evidence of floatables wash-up, fish kills, sediment accumulation, and
the frequency, duration, and magnitude of instream WQS violations).
Monitoring as part of the NMC is not intended to be extensive or costly. It should entail
collection of existing information from relevant agencies about the CSS, CSOs, the receiving water
body, and pollutant sources discharging to the same receiving waters, as well as preliminary
investigation activities such as field inspections and simple measurements using chalk boards, bottle
boards, and block tests. The collected information and data will be used to establish a baseline of
existing conditions for evaluating the efficacy of the technology-based controls and to develop the
LTCP (as described in Section 2.1.2).
Data analysis and field inspection activities also support implementation of several other
NMC:
l Proper operation and regular maintenance programs for the sewer system-
Characterization of the CSS will support the evaluation of the effectiveness of current
operation and maintenance (O&M) programs and help identify areas within the CSS that
need repair.
l Maximum use of the collection system for storage-Information gained during field
inspections, such as the system topography (e.g., location of any steep slopes) and the
need for regulator or pump adjustments, can assist in identifying locations where minor
modifications to the CSS can increase in-system storage.
l Dry weather overflows-Field inspections will assess the presence of dry weather
overflows, the conditions under which they occur, and the effectiveness of any control
measures in place.
The CSO Control Policy recommends that a Phase I permit require the permittee to develop
and submit an LTCP that, when implemented, will ultimately result in compliance with CWA
requirements. The permittee should use either the presumption approach or the demonstration
approach in developing an LTCP that will provide for WQS attainment. The two approaches are
discussed in more detail below and in Chapters 7 through 9.
The permittee should evaluate the data and information obtained through the initial system
characterization to determine which approach is more appropriate based on site-specific conditions.
Generally, the demonstration approach would be selected when sufficient data are available, or can
be collected, to “demonstrate” that a proposed LTCP is adequate to meet the water quality-based
requirements of the CWA. If sufficient data are not available and cannot be developed to allow use
of the demonstration approach, and the permitting authority believes it is likely that implementation
of a control program that meets certain performance criteria will result in attainment of CWA
requirements, the permittee would use the presumption approach.
"i. The planned control program is adequate to meet WQS andprotect designated uses,
unless WQS or uses cannot be met as a result of natural background conditions or
pollution sources other than CSOs;
ii. The CSO discharges remaining after implementation of the planned control program
will not preclude the attainment of WQS or the receiving waters ’ designated uses or
contribute to their impairment. Where WQS and designated uses are not met in part
because of natural background conditions or pollution sources other than CSOs, a
total maximum daily load, including a wasteload allocation and a load allocation,
or other means should be used to apportion pollutant loads;
iii. The planned control program will provide the maximum pollution reduction benefits
reasonably attainable; and
iv. The planned control program is designed to allow cost effective expansion or cost
effective retrofitting if additional controls are subsequently determined to be
necessary to meet WQS or designated uses. ” (Section II.C.4.b of the CSO Control
Policy)
Generally, monitoring and modeling activities will be integral to successfully demonstrating that
these criteria have been met.
Presumption Approach. This approach is based on the presumption that WQS will be
attained with implementation of an LTCP that meets certain performance-based criteria. For the
presumption approach, the CSO Control Policy states that:
“A program that meets any of the criteria listed below would be presumed to provide an
adequate level of control to meet the water quality-based requirements of the CWA, provided
the permitting authority determines that such presumption is reasonable in light of the data
and analysis conducted in the characterization, monitoring, and modeling of the system and
the consideration of sensitive areas described above. These criteria are provided because
data and modeling of wet weather events often do not give a clear picture of the level of CSO
controls necessary to protect WQS.
ii. The elimination or the capture for treatment of no less than 85% by volume of the
combined sewage collected in the CSS during precipitation events on a system-wide
annual average basis...
iii. The elimination or removal of no less than the mass of pollutants, identified as
causing water quality impairment..., for the volumes that would be eliminated or
capturedfor treatment under paragraph ii... ” (Section II.C.4.a.)
Monitoring and modeling activities are also likely to be necessary in order to obtain the permitting
authority’s approval for using the presumption approach. Considerations for using both the
presumption approach and the demonstration approach are discussed in Combined Sewer
Overflows - Guidance for Long Term Control Plan (U.S. EPA, 1995a).
Whether the LTCP ultimately reflects the demonstration approach or the presumption
approach, it should contain the same elements, as identified in the CSO Control Policy:
l Public participation
l Evaluation of alternatives
l Cost/performance considerations
l Operational plan
l Implementation schedule
Of these elements, the first and last are directly linked to monitoring and modeling and are
described below.
“In order to design a CSO control plan adequate to meet the requirements of the
CWA, a permittee should have a thorough understanding of its sewer system, the
response of the system to various precipitation events, the characteristics of the
overflows, and the water quality impacts that result from CSOs. The permittee
should adequately characterize through monitoring, modeling, and other means as
appropriate, for a range of storm events, the response of its sewer system to wet
weather events including the number, location and frequency of CSOs, volume,
concentration and mass ofpollutants discharged and the impacts of the CSOs on the
receiving waters and their designated uses. The permittee may need to consider
information on the contribution and importance of other pollution sources in order
to develop a final plan designed to meet water quality standards. The purpose of the
system characterization, monitoring and modeling program initially is to assist the
permittee in developing appropriate measures to implement the nine minimum
controls and, if necessary, to support development of the long-term CSO control
plan. The monitoring and modeling data also will be used to evaluate the expected
effectiveness of both the nine minimum controls and, if necessary, the long-term CSO
controls, to meet WQS. ” (Section II.C.1)
Characterization, monitoring, and modeling of the CSS can be broken into the following
elements:
Analysis of existing data should include an examination of rainfall records and available data
on flow, capacity, and water quality for the collection system, treatment plant, and receiving water.
This analysis, as well as information from field inspections and simple measurements, provides the
basis for the preliminary system characterization. This initial characterization of the system
(described in more detail in Chapter 3) should identify the number, location, and frequency of
overflows and clarify their relationship to sensitive areas, pollution sources within the collection
system (e.g., indirect discharges from nondomestic sources), other pollution sources discharging to
the receiving water (e.g., direct industrial discharges, POTWs, storm water discharges), and
background/upstream pollution sources (e.g., agricultural or other nonpoint source runoff).
Since some of these activities are also conducted as part of NMC implementation, the LTCP
should be developed in coordination with NMC implementation efforts. Ultimately, because the
LTCP is based on more detailed knowledge of the CSS and receiving waters than is necessary to
implement the NMC, the extent of monitoring and modeling for LTCP development is expected to
be more sophisticated.
Examination of existing data, field inspections and simple measurements, and other
preliminary characterization activities will serve as the basis for the development of a cost-effective
monitoring and modeling plan (discussed in Chapter 4). The monitoring and modeling plan should
be designed to provide the information and data needed to develop and evaluate CSO control
alternatives and to select cost-effective CSO controls.
Implementation of the monitoring and modeling plan should enable the permittee to predict
the CSS’s response to various wet weather events and evaluate CSO impacts on receiving waters for
alternative control strategies. Evaluation of CSO control alternatives is discussed in Combined
Sewer Overflows - Guidance for Long Term Control Plan (U.S. EPA, 1995a).
Based on the evaluation of control strategies, the permittee, in coordination with the public,
the NPDES permitting authority, and the State WQS authority, should select the cost-effective CSO
controls needed to provide for the attainment of WQS. Specific conditions relating to
implementation of these CSO controls will be incorporated into the NPDES permit as described in
Section 2.1.4.
The CSO Control Policy recommends that the Phase I permit require permittees to:
l
Develop and submit an LTCP as soon as practicable, but generally within two years after
permit issuance/modification.
The permittee should not view NMC implementation and LTCP development as independent
activities, but rather as related components in the CSO control planning process. Implementation
of the NMC establishes the baseline conditions upon which the LTCP will be developed.
In many cases, the LTCP will be developed concurrent with NMC implementation. As
described in Sections 2.1.1 and 2.1.2, both efforts require the permittee to develop a thorough
understanding of the CSS. For example, monitoring done as part of the NMC to effectively
characterize CSO impacts and the efficacy of CSO controls should provide a base of information and
data that the permittee can use in conducting more thorough characterization, monitoring, and
modeling activities for LTCP implementation.
Therefore, the characterization activities needed to implement the NMC and develop the
LTCP should be a single coordinated effort.
l Requirements for maximizing the treatment of wet weather flows at the treatment plant
l A reopener clause authorizing permit modifications if CSO controls fail to meet WQS
or protect designated uses.
The post-construction compliance monitoring program should provide sufficient data to determine
the effectiveness of CSO controls in attaining WQS. The frequency and type of monitoring in the
program will be site-specific. In most cases, some monitoring will be conducted during the
construction/implementation period to evaluate the effectiveness of the long-term CSO controls. In
some cases, however, it may be appropriate to delay implementation of the post-construction
monitoring program until construction is well underway or completed.
The post-construction compliance monitoring program may also include other appropriate
measures for determining the success of the CSO control program. Measures of success, which are
also discussed in Section 2.3, can address both CSO flow and quality issues. For example, flow-
related measures could include the number of dry weather overflows or CSO outfalls eliminated, and
reductions in the frequency and volume of CSOs. Quality-related measures could include decreases
in loadings of conventional and toxic pollutants in CSOs. Environmental measures focus on human
and ecosystem health trends such as reduced beach closures or fish kills, improved biological
integrity indices, and the full support of designated uses in receiving water bodies.
The watershed approach represents a holistic approach to understanding and addressing all
surface water, ground water, and habitat stressors within a geographically defined area, instead of
addressing individual pollutant sources in isolation. It serves as the basis for “place-based”
solutions to ecosystem protection.
Point and nonpoint source programs, the drinking water program, and other surface and
ground water programs are all integrated into the watershed approach. Under the watershed
approach, these programs address watershed problems in an effective and cooperative fashion. The
CSO Control Policy encourages NPDES permitting authorities to evaluate CSO control needs on
a watershed basis and coordinate CSO control program efforts with the efforts of other point and
nonpoint source control activities within the watershed.
The application of the watershed approach to a CSO control program is particularly timely
and appropriate since the ultimate goal of the CSO Control Policy is the development of long-term
CSO controls that will provide for the attainment of WQS. Since pollution sources other than CSOs
are likely to be discharging to the receiving water and affecting whether WQS are attained, the
permittee needs to consider and understand these sources in developing its LTCP. The permittee
should compile existing information and monitoring data on these sources from the NPDES
permitting authority, State watershed personnel, or even other permittees or dischargers within the
watershed. If other permittees within the watershed are also developing LTCPs, they may have an
opportunity to pursue a coordinated and cooperative approach to CSO control planning.
The sources of watershed pollution and impairment, in addition to CSOs, are varied and
include other point source discharges, discharges from storm drains, overland runoff, habitat
destruction, land use activities (such as agriculture and construction), erosion, and septic systems and
landfills. A watershed-based approach to LTCP development allows for the site-specific
determination of the relative impacts of CSOs and other pollution sources. The flows and loads from
the pollutant sources are estimated using available site-specific data and modeling. In addition to
locally available data, potential data sources include:
If the permittee determines during its LTCP development that WQS cannot be met because
of other pollution sources within the watershed, a total maximum daily load (TMDL), including
wasteload allocations for point sources and load allocations for nonpoint sources, may be necessary
to apportion loads among dischargers. Several publications provide TMDL and wasteload allocation
guidance (U.S. EPA, 1995g; U.S. EPA, 1991b; Mills et al., 1986; Mancini et al., 1983; Martin et al.,
1990; Mills et al., 1985a,b). In many cases, a TMDL may not have been developed for the
permittee’s watershed. In these cases, the monitoring and modeling conducted as part of the
development and implementation of long-term CSO controls will support an assessment of water
quality and could support the development of a TMDL. BASINS (U.S. EPA, 1997a) also supports
the development of TMDLs.
Of particular importance to CSO control planning and management is the NPDES Watershed
Strategy (U.S. EPA, 1994e). This strategy outlines national objectives and implementation activities
to integrate the NPDES program into the broader watershed protection approach. The Strategy also
supports the development of statewide basin management as part of an overall watershed
management approach. Statewide basin management is an overall framework for integrating and
coordinating water resource management efforts basin-by-basin throughout an entire State. This will
result in development and implementation of basin management plans that meet stated
environmental goals.
The Clean Water Action Plan, issued jointly by EPA and the U.S. Department of Agriculture,
calls for States to issue unified watershed assessments by October, 1998 (U.S. EPA/USDA, 1998).
Assessments identify degraded watersheds needing restoration, watersheds needing preventive action
to sustain water quality, and pristine or sensitive watersheds on Federal lands needing additional
protection. The Clean Water Action Plan identifies mechanisms for States and tribes to coordinate
with Federal agencies to prioritize watershed restoration and protection efforts. Additional
information is available at http://www.cleanwater.gov/.
Use of the comprehensive watershed approach during long-term CSO planning will promote
a more cost-effective program for achieving WQS in a watershed. LTCP development using the
watershed approach is discussed further in Combined Sewer Overflows - Guidance for Long-Term
Control Plan (U.S. EPA, 1995a).
Before developing a monitoring plan for characterizing the CSS and determining post-
construction compliance, the permittee should identify appropriate measures of success based on
site-specific conditions. Measures of success are objective, measurable, and quantifiable indicators
that illustrate trends and results over time. Measures of success generally fall into four categories:
l End-of-pipe measures that show trends in the discharge of CSS flows to the receiving
water body, such as reduction of pollutant loadings, the frequency of CSOs, and the
duration of CSOs
l Receiving water body measures that show trends of the conditions in the receiving water
body, such as trends in dissolved oxygen levels, sediment oxygen demand, and solids
and fecal coliform concentrations
l Ecological, human health, and use measures that show trends in conditions relating to
the use of the water body, its effect on the health of the population that uses the water
body, and the health of the organisms that reside in the water body, including beach
closures, attainment of designated uses, habitat improvements, and fish consumption
advisories. Such measures would be coordinated on a watershed basis as appropriate.
Measures of success for a CSO control program should typically include a balanced mix of measures
from each of the four categories.
As municipalities begin to collect data and information on CSOs and CSO impacts, they have
an important opportunity to establish a solid understanding of the “baseline” conditions and to
consider what information and data are necessary to evaluate and demonstrate the results of CSO
control. The permittee should choose measures of success that can be used to indicate reductions
in the occurrence and effects of CSOs. Municipalities and NPDES permitting authorities should
agree early in the planning stages on the data and information that will be used to measure success.
These measures of success may need to be adapted as a municipality gains additional information
during its system characterization. (Measures of success for the CSO program are discussed in
Combined Sewer Overflows-Guidance for Long-Term Control Plan (U.S. EPA, 1995a) and
Performance Measures for the National CSO Control Program (AMSA, 1996)). The permittee
should consider these measures of success when determining which parameters to include in its
monitoring plan.
The permittee may be subject to monitoring requirements for other regulated wet weather
discharges, such as storm water, in addition to CSOs. Due to the unpredictability of wet weather
discharges, monitoring of such discharges presents challenges similar to those for monitoring CSOs.
The permittee should coordinate all wet weather monitoring efforts. Developing one monitoring and
modeling program for all wet weather programs will enable the permittee to establish a clear set of
priorities for monitoring and modeling activities.
Section 301 of the CWA and NPDES regulations at 40 CFR 122.44 require the establishment
of both technology-based and water quality-based effluent limitations:
The development of permit limits and conditions for CSO permittees is described in greater
detail in Combined Sewer Overflows - Guidance for Permit Writers (U.S. EPA, 1995e).
Since CSO controls must ultimately provide for the attainment of WQS, the analysis of CSO
control alternatives should be tailored to the applicable WQS. A key principle of the CSO Control
Policy is the review and revision, as appropriate, of WQS and their implementation procedures to
reflect the site-specific wet weather impacts of CSOs. In identifying applicable WQS, the permittee
and the permitting and WQS authorities should consider whether revisions to WQS are appropriate
for wet weather conditions in the receiving water.
Review of WQS should be conducted concurrent with the development of the LTCP to
ensure that the long-term CSO controls will be sufficient to provide for the attainment of applicable
WQS. The information gained from LTCP development can then be used to support any efforts to
revise WQS. (The identification of applicable WQS and methods for assessing attainment of WQS
are discussed in Chapter 9).
The WQS program contains several types of mechanisms that could potentially be used to
address site-specific factors such as wet weather conditions. These include the following:
l Adopting partial uses to reflect situations where a significant storm event precludes the
use from occurring
l Adopting seasonal uses to reflect that certain uses do not occur during certain seasons
(e.g., swimming does not occur in winter)
l Defining a use with greater specificity (e.g., warm-water fishery in place of aquatic life
protection)
These potential revisions are described in detail in the Water Quality Standards Handbook,
Second Edition (U.S. EPA, 1994).
Reviewing and revising WQS requires the collection of information and data to support the
proposed revision. In general, a use attainability analysis (UAA) is required to support a proposed
WQS revision. The process for conducting UAAs for receiving waters has been described in various
EPA publications (U.S. EPA, 1994; U.S. EPA, 1984a; U.S. EPA, 1984b; U.S. EPA, 1983b).
The information and data collected during LTCP development could potentially be used to
support a UAA for a proposed revision to WQS to reflect wet weather conditions. Thus, it is
important for the permittee, NPDES permitting authority, State WQS authority, and EPA Regional
offices to agree on the data, information, and analyses that are necessary to support the development
of long-term CSO controls as well as the review of applicable WQS and implementation procedures,
if appropriate.
Development and implementation of a CSO monitoring and modeling program should not
be solely the permittee’s responsibility. Development of a successful and cost-effective monitoring
and modeling program should reflect the coordinated efforts of a team that includes the NPDES
permitting authority, State WQS authority, State watershed personnel, EPA or State monitoring
personnel, and any other appropriate entities.
l Determine, in coordination with the permittee and appropriate State and Federal
agencies, whether the permittee needs to consider any sensitive areas in developing a
monitoring and modeling plan
l Coordinate with the permittee to ensure that the monitoring requirements in the permit
are appropriately site-specific
l Assist in compiling relevant existing information, monitoring data, and studies at the
State and/or EPA Regional level
l Decide if the presumption approach is applicable based on the data and analysis
conducted in the characterization, monitoring, and modeling of the system and the
consideration of any sensitive areas
l Coordinate the permittee’s CSO monitoring and modeling efforts with monitoring and
modeling efforts of other permittees within the watershed
l Coordinate the team review of the monitoring and modeling plan, monitoring and
modeling data, and other components of the LTCP. To ensure team review of the
monitoring and modeling plan, the permitting authority could recommend that the plan
include a signature page for endorsement by all the team members after their review.
l Provide input on the review and possible revision of WQS, including conduct of a use
attainability analysis where necessary
l Assist in compiling existing State information, monitoring data, and studies for the
receiving water body
l Ensure that the permittee’s monitoring and modeling efforts are coordinated and
integrated with ongoing State monitoring programs
l Evaluate any special monitoring activities such as biological testing, sediment testing,
and whole effluent toxicity testing.
l Ensure that the permittee’s monitoring activities are coordinated with ongoing watershed
monitoring programs
l Assist in compiling existing State information, monitoring data, and studies for the
receiving water body
l Ensure the permittee’s monitoring and modeling efforts are integrated with TMDL
application or development.
l Assist in compiling relevant existing monitoring data and studies for the receiving water
body
l Provide information on available models and the monitoring data needed as model inputs
The public should also participate in development and implementation of the system
characterization activities and the monitoring and modeling program. Throughout the LTCP
development process, the public should have the opportunity to review and provide comments on
the results of the system characterization, monitoring, and modeling activities that lead to the
selection of long-term CSO controls. The public participation effort might involve public meetings
at key points during the system characterization phase of LTCP development. Input from the public,
obtained during the early phases of the planning process, will enable a municipality to better develop
an outreach program that reaches a broad base of citizens. In addition to public meetings,
municipalities can obtain input from telephone surveys, community leader interviews, and
workshops. Each of these activities can give the municipality a better understanding of the public
perspective on local water quality issues and sewer system problems, the amount of public concern
about CSOs in particular, and public willingness to participate in efforts to control CSOs.
The primary objectives of the existing data analysis and field investigation are:
l To determine the current level of understanding and knowledge of the CSS and receiving
water
l To identify the data that still need to be collected through the monitoring and modeling
program
The activities required to meet these objectives will vary widely from system to system.
Many permittees have already made significant progress in conducting initial system
characterizations. Implementation of the NMC, which was expected by January, 1997, should have
enabled permittees to compile a substantial amount of information on their CSSs. In addition,
studies by EPA, State agencies, or other organizations may provide substantial information and data
for the receiving water characterization.
This chapter describes the following activities in the initial system characterization:
The permittee should consult with the NPDES permitting authority and the review team (see
Section 2.6) when reviewing the results from the initial system characterization and in preparation
for developing the monitoring and modeling plan (Chapter 4). Performing and documenting initial
characterization activities may help satisfy certain requirements for NMC implementation and
documentation. Thus, it is essential that the permittee coordinate with the NPDES permitting
authority on an ongoing basis throughout the initial characterization process.
For the first part of the physical characterization, the permittee should compile, catalogue,
and review existing information on the design and construction of the CSS to evaluate how the CSS
operates, particularly in response to wet weather events. The permittee should compile, for the entire
CSS, information on the contributing drainage areas, the location and capacity of the POTW and
interceptor network, the location and operation of flow regulating structures, the location of all
known or suspected CSO outfalls, and the general hydraulic characteristics of the system (including
existing flow data for both wet weather and dry weather). Historical information is often available
from the following sources:
l
Sewer Maps of Suitable Scale- Sewer maps define the pipe network of the sewer
system and may indicate the drainage areas that contribute to each CSO outfall. Ideally,
they should include the combined, separate sanitary, and separate storm sewer systems,
manhole locations for monitoring access, catch basin locations, and pipe shapes and
materials. Sewer maps may also show curb/surface drainage, roof connections, pipe age,
and ongoing roadway construction projects and their influence on storm flow. Many
cities have also used Geographic Information Systems (GIS) to develop maps of their
sewer systems. Data provided from these maps, such as the invert elevations, can be
used to calculate individual pipe capacities and to develop detailed hydraulic models.
Sewer maps should be field checked because field conditions may differ significantly
from the plans (see System Field Investigations, Section 3.1.3).
l Topographic Maps- The U.S. Geological Survey (USGS) provides topographic maps,
usually with lo-foot contour intervals. The local municipality or planning agency may
have prepared topographic maps with finer contour intervals, which may be more useful
in identifying drainage areas contributing to CSOs.
l Aerial Photograph-When overlaid with sewer maps and topographic maps, aerial
photos may aid in identifying land uses in the drainage areas. Local planning agencies,
past land use studies, or State Departments of Transportation may have aerial
photographs suitable for the initial characterization.
l Diversion Structure Drawings- Drawings of CSS structures, in plan and section view,
indicate how the structures operate, how they should be monitored, and how they could
be altered to facilitate monitoring or improve flow control.
l
Rainfall Data- Rainfall data are one of the most important and useful types of data
collected during the initial system characterization. Reliable rainfall data are necessary
to understand the hydraulic response of the CSS and, where applicable, to model this
response. Sources of data may include long-term precipitation data collected from a
weather station within or outside the CSS drainage basin, or short-term, site-specific
precipitation data from stations within the drainage basin or sub-basins. Wastewater
treatment plants may also collect their own rainfall data or maintain records of rainfall
data from a local weather station.
Long-term rainfall data collected within the drainage basin provide the best record of
precipitation within the system and hence have the greatest value in correlating historic
overflow events with precipitation events and in predicting the likelihood of wet weather
events of varying intensities. If such data are not available, however, both long-term
regional and short-term local data may be used. For calibration and validation of
hydraulic models (see Section 7.4), it is important to use rainfall data collected from
within or in very close proximity to the drainage area.
National rainfall data are available from the National Weather Service, which operates
thousands of weather monitoring stations throughout the country. Rainfall data for some
areas are available on the Internet (the National Weather Service home page can be found
at http://www.nws.noaa.gov/). The local municipality, airports, universities, or other
State or Federal facilities can also provide rainfall data. The National Oceanic and
Atmospheric Administration (NOAA), National Climatic Data Center (NCDC), Climate
Services Branch is responsible for collecting precipitation data. Data on hourly, daily,
and monthly precipitation for each monitoring station (with latitude and longitude) can
be obtained on computer diskette, microfiche, or hard copy by calling (704) 259-0682,
or by writing to NCDC, Climate Services Branch, The Federal Building, Asheville, NC
28071-2733. Some NCDC data are also available on the Internet (NCDC’s home page
can be found at http://www.ncdc.noaa.gov/). The NCDC also provides a computer
program called SYNOP for data analysis.
Additionally, permittees with few or no rain gages located within the system drainage
basin may want to install one or more gages early in the CSO control planning process.
Collection and analysis of rainfall data are discussed in Chapters 4 and 5.
l GIS databases
l Treatment plant upgrade reports
l CSS flow records (for both dry weather and wet weather)
l Treatment plant and pump station flow and performance records
l Design specifications
l Infiltration/inflow (I/I) studies
l Sewer system evaluation surveys (SSES)
l Storm water master plans
The availability of these sources of information varies widely among permittees. Collection system
operation and maintenance personnel can be invaluable in determining the existence and location
of such data, as well as providing system knowledge and insight.
Using the historical data, the permittee should develop a map of the CSS, including the
drainage basin of combined sewer areas and separate storm sewer areas. Larger systems will find
it useful to map sub-basins for each regulating structure and CSO. This map will be used for
analyzing system flow directions and interconnections, analyzing land use and runoff parameters,
locating monitoring networks, and developing model inputs. The map can also be a valuable
planning tool in identifying areas of special concern in the CSS and planning further investigative
efforts and logistics. The map should be modified as necessary to reflect additional CSS and
receiving water information (such as the locations of other point source discharges to the receiving
water, the location of sensitive areas, and planned or existing monitoring locations), when these
become available.
l General land uses (e.g., residential, commercial, industrial) and degree of imperviousness
l Access points (e.g., manholes safely accessible considering traffic and pipe depth; flat,
open areas accessible for sampling)
l Pump stations
l River crossings
l Rain gages
l Existing monitoring locations (CSS, CSO, storm water, other point and nonpoint sources,
and receiving water)
l Soil types
l Outlying separate sanitary sewer areas draining to the CSS (where applicable)
l Other point source discharges such as industrial discharges and separate storm water
system discharges
It may be useful to generate two or more maps with different scales, such as a coarse-scale
map (e.g., 7.5-minute USGS map) for land uses and other watershed scale information and a finer-
scale map (e.g., 1” = 200’ or 1” = 400’) for sewer system details. In some cases, a Computer Aided
Design (CAD) or GIS approach can be used. Some advanced sewer models can draw information
directly from CAD tiles, eliminating the duplication of entering data into the model. A
municipality’s planning department may be a useful source for the hardware, software, and data
needed for such mapping efforts.
Before developing a monitoring and modeling program, the permittee should supplement
historical CSS information with field observations of the system to verify findings or fill data gaps.
For example, visual inspection of regulator chambers and overflow structures during dry and wet
weather verifies information included in drawings and provides data on current conditions. Further,
it is necessary to verify that gates or flow diversion structures operate correctly so that ensuing
monitoring programs collect information representative of the expected behavior of the system.
Field inspections should address all areas of the CSS, including the pipe network, flow diversion
structures, CSO outfalls, pump stations, manholes, and catch basins.
l Locate and clarify portions of the system not shown on as-built drawings
l Identify dry weather overflows and possible causes of the overflows (e.g., diversion
structures set too low)
l Examine the general conditions and operability of flow regulating equipment (e.g., weirs,
gates)
l Identify areas that are curbed, areas where roof downspouts are directly connected to the
CSS, and impervious areas.
Although generally beyond the scope of a small system characterization effort, in-line TV
cameras can be used to survey the system, locate connections, and identify needed repairs. WPCF
(1989) describes in-line inspection methods in detail and provides additional useful information for
system evaluations.
The field investigation may also involve preliminary collection of both dry weather and wet
weather flow and depth data, which can support the CSS flow monitoring and modeling activities
later in the CSO control planning process. Preliminary CSS flow and depth estimates can begin to
answer the following questions:
l How many dry weather overflows occur? How frequently and at which outfall(s)? How
much flow is being discharged during dry weather?
l How deep are the maximum flows at the flow diversion structures? Would alteration of
a diversion structure affect whether a CSO occurs?
A variety of simple flow measurement techniques can help answer these questions prior to
development and implementation of a monitoring and modeling plan. These include:
l Chalk Spraying- A sprayer is used to blow chalk into a CSO structure. Passing sewer
flow washes away the chalk, indicating approximate flow depth since spraying.
l Bottle Boards-A bottle board is a vertical board with a series of attached open bottles.
As flow rises the bottles with openings below the maximum flow are filled. When the
flow recedes the bottles remain full indicating the height of maximum flow
(see Exhibit 5-6).
l Block Tests-Block tests do not measure depth, but are used to detect the presence of an
overflow. A block of wood or other float is placed atop the overflow weir. If an
overflow occurs, it is washed off the weir indicating that the event took place. The block
can be tethered to the weir for retrieval.
These simple flow measurement techniques could be a useful component of the NMC for
monitoring to characterize CSO impacts and the efficacy of CSO controls. The permittee should
discuss this with the permitting authority. In some limited cases, automated continuous flow
monitoring may be used. These techniques and other CSS monitoring techniques are discussed in
Chapter 5.
The physical characterization of the CSS should include a flow balance, using a schematic
diagram of the collection system. Exhibit 3-1 provides an example of a basic flow balance diagram.
It shows expected wet weather and dry weather flows through each service area, and the likely flows
at each CSO based on sewer hydraulic capacities. The diagram can be expanded to include
additional detail, such as breaking down the cumulative flows at each regulator to show
schematically where the flows are entering the system. This can sometimes reveal local bottlenecks
that may be resolved by relocating the connection to a downstream portion of the system where there
is greater capacity.
The following steps can be used to develop a flow balance diagram or conduct a similar flow
analysis:
l Section the collection system into a series of basins of small enough area to characterize
the major collection system elements, differing land uses, receiving streams, and other
characteristics that may become important during the development of a monitoring and
modeling plan. These basins will likely be refined as work progresses.
l Establish the hydraulic capacity of each element of the system. For a preliminary
analysis, this can be done using the unsurcharged capacity of the system, based on pipe
size and slope, pump station capacity, and a knowledge of bottlenecks in the system.
River
* Cumulative flows = flows from the service area and service areas
upstream in the collection system. Wet weather flow values are for
the average of several sampled storm events.
l For each basin, develop a dry weather estimate of flow delivered to the system. This can
be done in a preliminary way by using total dry weather flow to the treatment plant,
disaggregated to each basin using population. Care should be taken where significant
differences in infiltration are suspected.
l For each basin, develop an estimate of wet weather inflow and wet weather-induced
infiltration. This estimate should be based on a consistent storm or return frequency in
each basin. (Flow monitoring in the CSS, including rainfall and runoff assessment, is
discussed in Chapter 5.)
l Display these data in a manner that aids data analysis, such as in a flow balance diagram
(Exhibit 3-1).
The schematic diagram, together with the historical data review and supplemental field study,
should enable the permittee to assign typical flows and maximum capacities to various interceptors
for non-surcharged flow conditions. Flow capacities can be approximated from sewer maps or
calculated from invert elevations. The resulting values provide a preliminary estimate of system
flows at peak capacity. Calculations of flow within intercepting devices or flow diversion structures
and flow records from the treatment plant help in locating sections of the CSS that limit the overall
hydraulic capacity.
The preliminary hydraulic analysis, together with other physical characterization activities,
will be useful in designing the CSS monitoring program and identifying areas that should receive
greater attention in developing the monitoring and modeling plan. This preliminary analysis can
help in identifying likely CSOs, the magnitude of rainfall that causes CSOs, estimated CSO volumes,
and potential control points. A hydraulic model may be useful in conducting the analysis.
As part of the initial system characterization, the permittee should review existing data to
determine the pollutant characteristics of combined sewage during both dry and wet weather
conditions, and, if possible, CSO pollutant loadings to the receiving water. The purpose of this effort
is to identify pollutants of concern in CSOs, their concentrations, and where possible, likely sources
of such pollutants. Together, these assessments will support decisions on what pollutants should be
monitored and where. This is discussed in detail in Chapter 4.
The POTW’s records can provide influent pollutant and flow data for both dry weather and
wet weather conditions. Such data can be analyzed to answer questions like:
l How do the influent volume, loads, and concentrations at the plant change during wet
weather?
l What is the average concentration of parameters such as solids, BOD, and metals at the
plant during wet weather flow?
l Which pollutants are discharged by industrial users, particularly significant industrial users?
For example, data analysis could include plotting a plant inflow time series by storm(s) and
comparing it to a rainfall time series plot for the same storm(s). In some cases, the permittee may
also be able to use POTW data to identify which portions of the CSS are contributing significant
pollutant loadings.
The permittee can potentially use national or regional storm water data (e.g., Nationwide
1
Urban Runoff Program (NURP) data ) (US. EPA, 1983a) to supplement its available data, although
more recent localized data are preferred. If approximate CSS flow volumes are known, approximate
CSS pollutant loads can be estimated using POTW data, CSS flow volume, and assumed storm water
concentration values. However, assumed constant or event mean concentration values for storm
water concentrations, such as NURP data, should be used with some reservation for CSOs since
concentrations vary during a storm and from storm to storm.
In order to obtain recent and reliable characterization data, the permittee may need to conduct
limited sampling at locations within the CSS as well as at selected CSO outfalls as part of the initial
system characterization. Since this limited sampling is usually less cost-effective than sampling
done as part of the overall monitoring program, the permittee should fully evaluate the need for such
data as part of the initial characterization. Chapter 5 provides details on CSS monitoring procedures.
3.2.2 Mapping
The permittee should plot existing pollutant characterization data on the study map for points
within the CSS as well as for CSO outfalls. This will highlight areas where no data exist and areas
with high concentrations of pollutants.
The third part of the initial system characterization is to establish the status of each receiving
water body impacted by CSOs. Using existing data and information and working with the NPDES
and water quality standards (WQS) authorities, the permittee should attempt to answer the following
types of questions:
1
Some NURP data may no longer be useful due to changed conditions (e.g., lead data might not apply since control
programs have been in place for many years). The permittee should contact the permitting authority to determine the
applicability of NURP data.
l Does the receiving water body contain sensitive areas (as defined by the CSO Control
Policy)?
l What are the applicable WQS? Is the receiving water body currently attaining WQS,
including designated uses?
l Are there particular problems in the receiving water body attributable wholly or in part
to CSOs?
l What are the hydraulic characteristics of the receiving water body (e.g., average flow,
tidal characteristics, instream flow regulations for dams and withdrawals)?
l What other dry and wet weather sources of pollutants in the watershed are discharging
to the receiving water body? What quantity of pollutants is being discharged by these
sources?
l What are the ecologic and aesthetic conditions of the receiving water body?
The following types of receiving water data will help answer these questions:
The permittee may already have collected receiving water data as part of other programs or
studies. For example, the NPDES permit may require sampling upstream and downstream of the
treatment plant outfall or the permittee may have performed special receiving water studies as part
of its NPDES permit reissuance process. Receiving water data may also be obtained through
consultation with the NPDES permitting authority, EPA Regional staff, State WQS personnel, and
State watershed personnel. The CWA requires States to generate and maintain data on certain water
bodies within their jurisdictions.
The following reports may provide information useful for characterizing a receiving water
body:
l
State 303(d) Lists- Under CWA section 303(d), States and authorized Tribes identify,
and establish total maximum daily loads (TMDLs) for, all waters that do not meet WQS
even after implementation of technology-based effluent limitations and any more
2
stringent effluent limitations or other pollution control requirements.
l
State 304(l) Lists- CWA section 304(l) required States to identify surface waters
adversely affected by toxic and conventional pollutants from point and non-point
sources, with priority given to waters adversely affected by point sources of toxic
3
pollutants. This one-time effort was completed in 1990. EPA recommends that the
permittee discuss with the permitting authority data on toxic “hot spots” identified under
this requirement.
State 305(b) Reports- Under CWA section 305(b), States must submit a water quality
assessment report to EPA every two years.
Section 319 State Assessment Reports- Under CWA section 319, States were required
to identify surface waters adversely affected by nonpoint sources of pollution, in a one-
time effort following enactment of the 1987 CWA Amendments.
Generally, permittees may retrieve this information at EPA or State offices, EPA’s Storage
and Retrieval of U.S. Waterways Parametric Data (STORET) system, EPA’s Water Quality System
resident within STORET, or EPA’s Water Body System (WBS). Since these data bases might not
include the particular water bodies being evaluated, the permittee should contact State officials prior
to seeking the data.
2
EPA recommends that the permittee discuss with the permitting authority the status of existing TMDL reports and
the schedule for doing new TMDLs for the CSO-impacted receiving water bodies.
3
These lists are not complete for some locations, so the lists should be discussed with State WQS staff before they
are used extensively.
In addition, studies conducted under enforcement actions, new permitting actions, and special
programs and initiatives may provide relevant data on receiving water flow, quality, and uses.
BASINS (Better Assessment Science Integrating Point and Nonpoint Sources) contains water quality
monitoring data and data on point sources and land use (US. EPA, 1997a). EPA’s EMAP
(Environmental Monitoring and Assessment Program) contains data on a limited number of
receiving waters and the EMAP Internet site (http://www.epa.gov/emap/) provides links to other
sources of environmental data (including STORET). EPA and State personnel may have information
on studies conducted by other Federal organizations, such as the U.S. Fish and Wildlife Service, the
U.S. Army Corps of Engineers, USGS, and the National Biological Service, and other organizations
such as The Nature Conservancy and formalized volunteer groups. For example, USGS’s National
Water-Quality Assessment (NAWQA) Program contains water quality information on 60 U.S. river
4
basins and aquifers. The permittee may save considerable time and expense by consulting directly
with these entities during the initial system characterization.
The receiving water characterization should also include an evaluation of whether CSOs
5
discharge to sensitive areas, which are a high priority under the CSO Control Policy. The LTCP
should prohibit new or significantly increased overflows to sensitive areas and eliminate or relocate
such overflows wherever physically possible and economically achievable. (This is discussed in
more detail in Combined Sewer Overflows - Guidance for Long-Term Control Plan, U.S. EPA,
1995a). The permittee should work with the NPDES permitting authority, the U.S. Fish and Wildlife
Service, and relevant State agencies to determine whether particular receiving water segments may
be considered sensitive under the CSO Control Policy.
In addition to reviewing existing data, the permittee may wish to conduct an observational
study of the receiving water body, noting differences in depth or width, tributaries, circulation (for
4
Information on the NAWQA program is available from USGS (703-648-5716) and the USGS Internet site
(http://wwwrvares.er.usgs.gov/nawqa/).
5
Sensitive areas, as discussed in the CSO Policy, are defined by the NPDES authority but include Outstanding
National Resource Waters, National Marine Sanctuaries, waters with threatened or endangered species and their habitat,
waters with primary contact recreation, public drinking water intakes or their designated protection areas, and shellfish
beds.
estuaries), point sources, suspected nonpoint sources, plant growth, riparian zones, and other
noticeable features. This information can be used later to define segments for a receiving water
model.
To supplement the observational study, the permittee may consider limited chemical or
biological sampling of the receiving water. Biocriteria or indices may be used in States such as Ohio
that have systems in place. Biocriteria describe the biological integrity of aquatic communities in
unimpaired waters for a particular designated aquatic life use. Biocriteria can be numerical values
or narrative conditions and serve as a reference point since biological communities in the unimpaired
waters represent the best attainable conditions (U.S. EPA, 1991 a). A limitation of biocriteria is that
they normally do not take into account wet weather conditions unique to urban streams, such as
runoff from highly impervious areas.
3.3.2 Mapping
The permittee should plot existing receiving water characterization data on the study map.
This will permit visual identification of areas for which no data exist, potential areas of concern, and
potential monitoring locations. GIS mapping can be used as an aid in this process. In addition to
the elements listed in Section 3.1.2 and 3.2.2, the map could include the following:
l WQS classifications for receiving waters at discharge locations and for upstream and
downstream reaches, and an indication of whether receiving waters are tidal or non-tidal
l Location of sensitive areas such as downstream beaches, other public access areas,
drinking water intakes, endangered species habitats, sensitive biological populations or
habitats, and shellfishing areas
l Locations of structures, such as weirs and dams, that can affect pollutant concentrations
in the receiving water
l Locations of access points, such as bridges, dams, and existing monitoring stations (such
as USGS stations), that make convenient sampling sites.
The final task in the initial system characterization is to identify gaps in information that is
essential to a basic understanding of the CSS’s response to rain events and the impact of CSOs on
the receiving water. The following questions may help to identify data gaps that need to be
addressed in the monitoring and modeling plan:
l Is sufficient information on the location, size, and characteristics of the sewers available
to support more complex analysis, including hydraulic modeling (as needed)?
l Are the minimum amount of rainfall and minimum rainfall intensity that cause CSOs at
various outfalls known?
l Are there differences between POTW wet weather and dry weather operations? If so, are
these clearly understood? (Improved wet weather operation can increase capture of CSS
flows significantly.)
l Is sufficient information available on pollutant loadings from CSOs and other sources
to support an evaluation of long-term CSO control alternatives?
l Are locations of sensitive areas and designated uses identified on a study map?
l Have existing monitoring locations in the receiving water been identified? Have
potential monitoring locations (e.g., safe, accessible points) in the receiving water been
identified for areas of concern and areas where no data exist?
l Are sufficient data available to assess existing water quality problems and the potential
for future water quality problems, including information on:
- Streambank erosion
- Sediment accumulation
- Dissolved oxygen levels
- Bacterial problems, such as those leading to beach closures
- Toxicity (metals)
- Nuisance algal or aquatic plant growths
- Damage to a fishery (e.g., shellfish beds)
- Damage to a biological community (e.g., benthic organisms)
- Floatables or other aesthetic concerns?
The answers to these types of questions will support the development of goals and objectives for the
monitoring plan, as described in Chapter 4.
Under the CSO Control Policy, the permittee should begin immediately to characterize its
combined sewer system (CSS), document implementation of the nine minimum controls (NMC),
and develop a long-term control plan (LTCP). The NMC and the LTCP both contain elements that
involve monitoring and modeling activities. The NMC include monitoring to characterize CSO
impacts and the efficacy of CSO controls, while the LTCP includes elements for characterization,
monitoring, and modeling of the CSS and receiving waters, evaluation and selection of CSO control
alternatives, and development of a post-construction monitoring program. As discussed in
Chapters 2 and 3, “monitoring” as part of the NMC involves gathering and analyzing existing data
and performing field investigations, but does not generally involve sampling or the use of complex
models. Thus the monitoring and modeling elements discussed in this chapter and subsequent
chapters primarily pertain to LTCP development and implementation.
The NPDES permit is likely to contain requirements for monitoring necessary to develop and
implement an LTCP. In many cases, the permit will first require the permittee to submit a
monitoring and modeling plan. For example, the Phase I permit may require submission of a
monitoring and modeling plan as an interim deliverable during LTCP development.
A well-developed monitoring and modeling plan is essential throughout the CSO planning
process to provide useful monitoring data for system characterization, evaluation and selection of
control alternatives, and post-construction compliance monitoring. Development of the plan is likely
to be an iterative process, with changes made as more knowledge about the CSS and CSOs is gained.
The permittee should aggressively seek to involve the NPDES permitting authority, as well as State
water quality standards (WQS) personnel, State watershed personnel, and EPA Regional staff,
throughout this process.
This chapter describes how the permittee can develop a monitoring and modeling plan that
provides essential and accurate information about the CSS and CSOs, and the impact of CSOs on
the receiving water. The chapter discusses the identification of monitoring and modeling goals and
objectives and the development of a monitoring and modeling plan to achieve those goals and
objectives. It provides detailed discussions and examples on identifying sampling locations,
frequencies, and parameters to be assessed. In addition, it briefly discusses certain monitoring and
modeling plan elements that are common to all system components being monitored. Readers
should consult the appropriate EPA guidance documents (see References) for further information
on topics such as chain-of-custody, sample handling, equipment, resources, and quality
assurance/quality control (QA/QC) procedures.
A monitoring and modeling plan can be developed with the following steps:
Step 1: Define the short- and long-term objectives - In order to identify wet weather
impacts and make sound decisions on CSO controls, the permittee should first formulate the short-
and long-term objectives of the monitoring and modeling effort. Every activity proposed in the plan
should contribute to attaining those objectives. (Step 1 is discussed in Section 4.1.1.)
Step 2: Decide whether to use a model - The permittee should decide whether to use a
model during LTCP development (and, if so, which model to use). This decision should be based
on site-specific considerations (e.g., CSS characteristics and complexity, type of receiving water)
and the information compiled in the initial system characterization. If a permittee decides to use a
model, the monitoring and modeling plan should include a modeling strategy. (Section 4.1.2)
Step 3: Identify data needed - The permittee should identify the monitoring data needed
to meet the goals and objectives. If modeling is planned, the monitoring plan should include any
additional data needed for model inputs. (Section 4.1.3)
Step 4: Identify sampling criteria (e.g., locations, frequency) - The permittee should
identify monitoring locations within the CSS, which CSOs to monitor, and sampling points within
the receiving water body. The permittee must also determine the frequency and duration of
sampling, parameters to be sampled, appropriate sample types to be collected (e.g., grab, composite),
and proper sample handling and preservation procedures. If a model will be used, the monitoring
plan should include any additional sampling locations, sample types, and parameters necessary to
adequately support the proposed model. If this is not feasible, the permittee may need to reevaluate
the model choice and select a different or less-complex model. (Sections 4.2 to 4.7)
Step 5: Develop data management and analysis procedures - A monitoring and modeling
plan also needs to specify QA/QC procedures and a data management program to facilitate storage,
use, and analysis of the data. (Section 4.8)
Step 6: Address implementation issues - Finally, the monitoring and modeling plan should
address implementation issues, such as record keeping and reporting, responsible personnel,
scheduling, and the equipment and resources necessary to accomplish the monitoring and modeling.
(Section 4.9)
The ultimate goal of a CSO control program is to implement cost-effective controls to reduce
water quality impacts from CSOs and provide for compliance with CWA requirements, including
attainment of WQS. Monitoring and modeling will foster attainment of this goal by generating data
to support decisions for selecting CSO controls. The monitoring and modeling plan should identify
how data will be collected and used to meet the following goals:
- What improvements in water quality will result from proposed CSO control
alternatives in the LTCP?
- How will the CSS hydraulics and CSO frequency and duration change under various
control alternatives?
- What is the best combination of control technologies across the system?
- Can CSO flows to sensitive areas be eliminated? If not, can they be relocated to less
sensitive areas?
In addition to selecting and implementing long-term CSO controls, the permittee will also
be required to develop and implement a post-construction compliance monitoring program. For this
type of monitoring program, the goal will typically be to:
Besides the broad goals, a municipality may have some site-specific objectives for its
monitoring program. For example, a permittee that is considering sewer separation as a CSO control
alternative may wish to assess the likely impacts of increased storm water loads on receiving waters.
The permittee should distinguish between short-term and long-term monitoring objectives.
Determining the length of short-term and long-term planning horizons will depend in part on how
much CSO control is already in place.
In developing a monitoring and modeling plan, the permittee should consider up front
whether to use modeling. If a permittee has a relatively simple system with a limited number of
outfalls, the use of flow balance diagrams and similar analyses may be sufficient and modeling may
not be necessary. For more complex systems, modeling can help characterize and predict:
Modeling also assists in formulating and testing the cause-effect relationships between wet
weather events and receiving water impacts. This knowledge can help the permittee evaluate control
alternatives and formulate an acceptable LTCP. Modeling enables the permittee to predict the
effectiveness of a range of potential control alternatives. By assessing the expected outcomes of
control alternatives before their implementation, the permittee can make more cost-effective
decisions. Modeling results may also be relevant to reviewing and revising State WQS. Since the
use of a model and its level of complexity affect the need for monitoring data, the permittee should
determine early on whether modeling is needed to provide sufficient information for making CSO
control decisions.
l Predict CSO occurrence, volume, and in some cases, pollutant characteristics, for rain
events other than those that occurred during the monitoring phase. These can include a
storm event of large magnitude (with a long recurrence period) or numerous storm events
over an extended period of time.
l Predict the wet weather performance of portions of the CSS that have not been monitored
extensively.
l Develop CSO statistics such as annual number of CSOs and percent of combined sewage
captured (particularly useful for municipalities pursuing the presumption approach under
the CSO Control Policy).
l Optimize sewer system performance as part of the NMC. In particular, modeling can
assist in locating storage opportunities and hydraulic bottlenecks and demonstrate that
system storage and flow to the POTW are maximized.
l Evaluate and optimize control alternatives, from simple controls described under the
NMC (such as raising weir heights to increase in-line storage) to more complex controls
proposed in the LTCP. The model can be used to evaluate the resulting reductions in
CSO volume and frequency.
l To predict the number and duration of WQS exceedances in areas of interest (such as
beaches or other sensitive areas).
If the permittee decides to model, the monitoring and modeling plan should include a
modeling strategy. There are several considerations in developing an appropriate modeling strategy:
l Meeting the expectations of the CSO Policy- The focus of modeling depends in part
on whether the permittee adopts the presumption or demonstration approach under the
CSO Policy. For some communities, the demonstration approach can necessitate
detailed simulation of receiving water impacts to show that CWA requirements will be
met under selected CSO control measures. The presumption approach may not involve
as much receiving water modeling since it presumes that CWA requirements are met
based on certain performance criteria, such as the maximum number of CSO events or
the percent capture of flows entering the system during a wet weather event.
l Meeting information needs at optimal cost- The modeling strategy should identify
modeling activities that provide answers as detailed and accurate as needed at the lowest
corresponding expense and effort. Since more detailed, accurate models are more
difficult and expensive to use, the permittee needs to identify the point at which an
increased modeling effort would provide diminishing returns. The permittee may use an
incremental approach, initially using simple screening models with limited data. These
results may then lead to refinements in the monitoring and modeling plan so that the
appropriate data are generated for more detailed modeling. Another option is to use a
simpler CSS model for the whole system and selectively apply a more complex sewer
model to portions of the system to answer specific design questions.
The monitoring effort necessary to address each goal will depend on a number of factors:
the layout of the collection system; the quantity, quality, and variability of the existing historical data
and the necessary additional data; whether modeling will be done and, if so, the complexity of the
selected model; and the available budget. In some cases, the initial characterization will yield
sufficient historical data so that only limited additional monitoring will be necessary. In other cases,
considerable effort may be necessary to fully investigate the characteristics of the CSS, CSOs, and
receiving waters. Some municipalities may choose to allocate a relatively large portion of the
available budget to monitoring, while others may allocate less. Because data needs may change as
additional knowledge is obtained, the monitoring program must be a dynamic program that evolves
to reflect any changes in data needs.
In addition to identifying the goals and objectives, the monitoring and modeling plan should
generally contain the following major elements:
l
Development of Sampling Program to Address Data Needs (discussed in Chapters 4-6)
l
Discussion of Methods for Data Management and Analyses (discussed in Chapters 4-9)
The checklists in Appendix A, Tables A-l and A-2 list items that should be addressed in
formulating a monitoring program. Elements in the first checklist should be part of any monitoring
program and cover seven major areas: sample and field data collection, laboratory analysis, data
management, data analysis, reporting, information use, and general. The second checklist applies
specifically to CSO monitoring and covers three areas: mapping of the CSS and identification of
monitoring locations, monitoring of CSO volume, and monitoring of CSO quality.
Because each permittee’s CSS, CSOs, and receiving water body are unique, it is not possible
to recommend a generic, “one-size-fits-all” monitoring and modeling plan in this document. Rather,
each permittee should design a cost-effective monitoring and modeling plan tailored to local
conditions and reflecting the size of the CSS, the impacts of CSOs, and whether modeling will be
performed. It should balance the costs of monitoring against the amount of data and information
needed to develop, implement, and verify the effectiveness of CSO controls.
While a monitoring and modeling budget may initially seem large, it is often a small
percentage of the total cost of CSO control. Each municipality should balance the cost of monitoring
and modeling against the risk of developing ineffective or unnecessary CSO controls based on
insufficient or inaccurate data. The information obtained from additional monitoring and modeling
may very well be offset by the reduction in total CSO costs.
The duration of the monitoring program will vary from location to location and reflect the
number of storm events needed to provide the data for calibrating and validating the CSS hydraulic
model (if a model is used), and evaluating CSO control alternatives and receiving water impacts.
During that period (which generally may be a season or several months), the permittee should
monitor storms of varying intensity, antecedent dry days, and total volume to ensure that calculations
and models represent the range of conditions experienced by the CSS.
The monitoring program should span enough storm events to enable the permittee to fully
understand the pollutant loads from CSOs, including the means and variations of pollutant
concentrations and the resulting effects on receiving water quality. If the permittee monitors only
a few storm events, the analysis should include appropriately conservative assumptions because of
the uncertainty associated with small sample sizes. For example, if monitoring data are collected
from a few storms during spring, when CSOs are generally larger and more frequent, mean pollutant
concentrations may be lower due to dilution from snowmelt and heavier rainfall and diminished first-
flush effects. When monitoring data are collected for additional storms, including those in the
summer and fall when CSOs are less frequent, the mean pollution concentrations may increase
significantly. Additional samples should reduce the level of uncertainty and allow the use of a
smaller margin of safety in the analysis.
The value of additional monitoring diminishes when additional data would result in a limited
change in the estimated mean and variance of a data set. The permittee should assess the value of
additional data as they are collected by reviewing how the estimated mean and variance of
contaminant concentrations changes over time. If estimated values stabilize (i.e., the mean and
variance show almost no change as additional monitoring results are added to the data set), the need
for additional data should be reassessed.
Pollutant loadings vary according to the number of days since the last storm and the intensity
of previous rainfalls. Therefore, to better represent the variability of actual conditions, the
monitoring program should be designed to sample storms with a variety of pre-storm conditions.
The monitoring and modeling plan should describe the sampling and analytical procedures
that will be used. Sample types depend on the parameter, site conditions, and the intended use of
the data. Flow-weighted composites may be most appropriate for determining average loadings of
pollutants to the receiving stream. Grab samples may suffice if only approximate pollutant levels
are needed or if worst-case conditions (e.g., first 15 or 30 minutes of overflow) are being assessed.
In addition, grab samples should be collected for pollutant parameters that cannot be cornposited,
such as oil and grease, pH, and bacteria. The monitoring plan should follow the sampling and
analytical procedures in 40 CFR Part 136, including the use of appropriate sample containers, sample
preservation methods, maximum allowable holding times, and analytical methods referencing one
or more of the following:
l Test methods in Appendix A to 40 CFR Part 136 (Methods for Organic Chemical
Analysis of Municipal and Industrial Wastewater)
l Standard Methods for the Analysis of Water and Wastewater (use the most current, EPA-
approved edition)
l Methods for the Chemical Analysis of Water and Wastes (U.S. EPA, 1979.
EPA 600/4-79-020).
In some cases, other well-documented analytical protocols may be more appropriate for
assessing in-stream parameters. For example, in estuarine areas, a protocol from NOAA’s Status
and Trends Program may provide better accuracy and precision if it reduces saltwater interferences.
To satisfy the objectives of the CSO Control Policy, the monitoring and modeling plan
should specify how the CSS and CSOs will be monitored, including monitoring locations,
frequencies, and pollutant parameters. The plan should be coordinated with other concurrent
sampling efforts (e.g., ongoing State water quality monitoring programs) to reduce sampling and
monitoring costs and maximize use of available resources. Careful selection of monitoring locations
can minimize the number of monitors and monitoring stations needed.
The monitoring and modeling plan should specify how rainfall data, flow data, and pollutant
data will be collected to define the CSS’s hydraulic response to wet weather events and to measure
CSO flows and pollutant loadings. The monitoring program should also provide background data
on conditions in the CSS during dry weather conditions, if this information is not already available
(see Chapter 3). Dry weather monitoring of the CSS may help identify pollutants of concern in
CSOs during wet weather.
Recommended spacing is the subject of a variety of research papers. The CSO Pollution
Abatement Manual of Practice (WPCF, 1989) provides the following summary of recommendations
on rain gage spacing:
“In Canada, rainfall and collection system modelers recommend one gauge every
1 or 2 kilometers. In Britain, the Water Research Center has recommended only half
that density, or one gauge every 2 to 5 kilometers. In the United States current
spacing recommendations are related to thunderstorm size. The average
thunderstorm is 6 to 8 kilometers in diameter,.. Therefore rain gauges are frequently
spaced every 6 to 8 kilometers . . . ”
For small watersheds, rain gages may need to be placed more closely than every 6 to 8 kilometers
so that sufficient data are available for analysis and model calibration. The monitoring and modeling
plan should document the rationale for rain gage spacing. Additional gages can provide valuable
information for CSS analysis and modeling and are usually a relatively inexpensive investment.
To obtain complete flow and pollutant loading data, the plan should also target portions of
the collection system that are likely to receive significant pollutant loadings. The plan should
identify locations where industrial users discharge into the collection system, and specify any
additional monitoring that will be conducted to supplement data collected through the industrial
pretreatment program. The plan should give special consideration to these areas when they are
located near CSO outfalls. Section 4.3.3 discusses the types of pollutants to be monitored.
Even if a monitoring program accounts for most of the total land area or estimated runoff,
monitoring other outfall locations, even with simple techniques, can provide information about
problem areas. For example, at an overflow point with only 10 percent of the contributing drainage
area, a malfunctioning regulator may result in discharges during dry weather or during small storms
when the interceptor has remaining capacity. As a result, this overflow point may become a major
contributor of flows. A simple technique such as a block test could identify this problem.
Alternatively, flow measurement equipment can be rotated between locations so that some
locations are monitored for a subset of the storms studied. For example, during one storm the
permittee could monitor critical outfalls with automated flow monitoring equipment, two less-
important outfalls with portable flow meters, and the others using chalk boards. During a second
storm, the permittee could still monitor critical outfalls with automated flow equipment but rotate
the portable flow meters to two other outfalls of secondary importance. However, since variability
is usually greater from storm to storm than from site to site, it is generally preferable to measure
more storms at a set of representative sampling sites than to rotate between all CSO locations.
If it is not feasible to monitor all outfalls, the permittee should identify a specific percentage
of the outfalls to be monitored based on the size of the collection system, the total number of
outfalls, the number of different receiving water bodies, and potential and known impacts. The
selected locations should represent the system as a whole or represent the worst-case scenario (for
example, where overflows occur most frequently, have the largest pollutant loading or flow volume,
or discharge to sensitive areas). If a representative set of CSO locations is selected for monitoring,
the results can be more easily extrapolated to non-monitored areas in the system.
l Drainage Area Flow Contribution- The relative flow contributions from different
drainage areas can be used to prioritize flow and pollutant monitoring efforts. There are
several methods for estimating relative flow contributions. The land area of each
outfall’s sub-basin provides only an approximate estimate of the relative flow
contribution because regulator operation and land use characteristics affect overflow
1
volume. Other estimation methods, such as the rational method , account for the runoff
characteristics of the upstream land area and produce relative peak flows of individual
drainage areas. Flow estimation using Manning’s equation (see Section 5.3.1) may
produce a better estimate of the relative flow contribution by drainage area.
l Land Use- During the initial sampling effort, the permittee should estimate the relative
contribution of pollutant loadings from individual drainage areas. Maps developed
during the initial system characterization should provide land use information that can
be used to derive pollutant concentrations for the different land uses from localized data
bases (based on measurements in the CSS). If local data are not available, the permittee
may use regional land use-based National Urban Runoff Program (NURP) studies,
although NURP data reflect only storm water and must be adjusted for the presence of
sanitary sewage flows and industrial wastewater. Pollutant concentration and drainage
area flow data can then be used to estimate loadings. Since pollutant concentrations can
vary greatly for different land uses, monitoring locations should represent subdivisions
of the drainage area with differing land uses.
1
The rational method is described in Schwab, et al., 1981.
January 1999
Chapter 4 Monitoring and Modeling Plan
l Location of Sensitive Areas- Since the LTCP should give the highest priority to
controlling overflows to sensitive areas, the monitoring and modeling plan should
identify locations where CSOs to sensitive areas, and their impacts, will be monitored.
l
Feasibility and Safety of Using the Location- After using the above criteria to identify
which outfalls will provide the most useful data, the permittee should determine whether
the locations are safe and accessible and identify which safety precautions are necessary.
If it is not feasible or practical to monitor at the point of discharge, the permittee should
select the closest upstream or downstream location that is still representative of the
overflow.
Example 4-1 illustrates one approach to selecting discharge monitoring sites for a
hypothetical CSS with ten outfalls. The selected outfalls-1, 4, 5, 7, and 9- discharge flow from
more than 60 percent of the total drainage area and 70 percent of the industrial area. Outfalls 1 and
5 are adjacent to sensitive areas. These five outfalls should provide sufficient in-depth coverage for
the city’s monitoring program. Simplified flow and modeling techniques at outfalls 2, 3,6, 8, and
10 can supplement the collected monitoring data and allow estimation of total CSS flow.
Combined Sewer Overflows - Guidance for Screening and Ranking (U.S. EPA, 1995c)
provides additional guidance on prioritizing monitoring locations. Although generally intended for
ranking CSSs with respect to one another, the techniques in this reference may prove useful for
ranking outfalls within a single system.
The permittee should monitor a sufficient number of storms to accurately predict the CSS’s
response to rainfall events and the characteristics of resulting CSOs. The frequency of monitoring
should be based on site-specific considerations such as CSO frequency and duration, which depend
on the rainfall pattern, antecedent dry period, type of receiving water and circulation pattern or flow,
ambient tide or stage of river or stream, and diurnal flow to the treatment plant.
A municipality has a combined sewer area with 4,800 acres and 10 outfalls discharging into a large river.
Exhibit 4-1 shows the characteristics of the discharge points that are potentially useful in choosing which
intercepting devices to monitor. Investigators used sewer and topographic maps to determine the size of the
drainage areas. Aerial photographs and information from a previous study indicated land use. Sewer maps,
spot checked in the field, verified the type of regulating structure. The sewer map and discussions with CSS
personnel provided information about safety and ease of access.
Outfalls 7 and 9 account for 33 percent of the total drainage area, and monitoring at outfall 7 would provide
data on commercial and industrial land uses that may have relatively higher pollutant loadings. These sites
pose no safety/accessibility concerns, making them desirable sampling locations.
Outfall 5 discharges in an area that is predominantly residential and includes one of the largest parks in the
municipality. This park has many recreational uses, including swimming during the warmer months. Since
areas used for primary contact recreation are considered sensitive areas, they are given highest priority in the
permittee’s LTCP under the CSO Control Policy. This outfall, which accounts for about 10 percent of the
drainage area, should be monitored.
Outfall 4, which is served by a pump station, accounts for 8 percent of the discharge area and includes
commercial areas. At this outfall, a counter or timer on the pump contacts or the use of full pipe flow
measurement devices usually provides an accurate measure of flow.
Outfall 1 discharges near the north edge of town, just before the river curves at its entrance to the
municipality. This outfall is located near a portion of the river that serves as a threatened species habitat and
therefore is considered a sensitive area. Since sensitive areas should be given the highest priority, this outfall
will be monitored. Monitoring this outfall also accounts for 13 percent of the total drainage area and a
significant portion of the area with commercial land uses.
In total, these five outfalls account for approximately 64 percent of the drainage area and more than 70
percent of the industrial land use.
The remaining sites pose practical problems for monitoring. Outfall 3 is difficult to access and poses safety
concerns. Outfalls 2, 6, 8, and 10 all have backwater effects, and access/safety concerns further limit
monitoring opportunities.
l
Outfall 2- Backwater effects, difficult access rating and safety concerns
l Outfall 3- Residential drainage area similar to Outfall 5, but difficult access rating and safety concerns
l
Outfall 6- Large residential drainage area but backwater effects and access/safety concerns limit
monitoring opportunities
l
Outfall 8- Drainage area small, but includes industrial and commercial land uses. Backwater effects
and access/safety concerns limit monitoring opportunities
l
Outfall 10- Backwater and difficult access limit monitoring opportunities.
l A certain number of precipitation events (e.g., monitor until five storms are
sampled-each storm may need to meet a certain minimum size)
A range of storm sizes should be sampled, if possible, to characterize the CSS response for
the variety of storm conditions that can occur. These data can be useful for long-term simulations.
Section 4.6 discusses a strategy for determining whether to monitor a particular wet weather event.
Overall, more frequency monitoring is warranted where:
l CSOs discharge to sensitive or high-quality areas, such as waters with drinking water
intakes or swimming, boating, and other recreational activities
l CSO flow volumes per inch of rainfall vary significantly from storm event to storm
event.
The number of samples collected will also reflect the type of sample collected. Where
possible, the permittee should collect flow-weighted composite samples to determine the average
pollutant concentration over a storm event (also known as the event mean concentration or EMC).
This approach decreases the analytical cost of a program based on discrete samples. Certain
parameters, such as oil and grease and bacteria, however, have limited holding times and must be
collected by grab sample (see discussion in Section 5.4.1). Also, when the permittee needs to
determine whether a pattern of pollutant concentration, such as a first-flush phenomenon, occurs
during storms, the monitoring program should collect several samples from the same locations
throughout a storm.
permittee should carefully consider the tradeoffs involved in committing resources to a sampling
program. A small number of samples may necessitate more conservative assumptions or result in
more uncertain assumptions because of high sample variability. A larger data set might better
determine pollutant concentrations and result in a more detailed analysis, enabling the permittee to
optimize any investment in long-term CSO controls. On the other hand, a permittee should avoid
spending large sums of money on monitoring when the additional data will not significantly enhance
the permittee’s understanding of CSOs, CSO impacts, and design of CSO controls. The permittee
should work closely with the NPDES permitting authority and the review team to design a
monitoring program that will adequately characterize the CSS, CSO impacts on the receiving water
body, and effectiveness of proposed CSO control alternatives.
The monitoring and modeling plan should state how the permittee will determine the
concentrations of pollutants carried in the combined sewage and the variability of these
concentrations during a storm, from outfall to outfall, and from storm to storm. Pollutant
concentration data should be used with flow data to compute pollutant loadings to receiving waters.
In some cases such data can also be used to detect the sources of pollutants in the system.
The monitoring and modeling plan should identify which parameters will be monitored.
These should include pollutants with water quality criteria for the specific designated use(s) of the
receiving water. The NPDES permitting authority may have specific guidance regarding parameters
for CSO monitoring. Parameters of concern may include:
2
Concentrations of bacteria in CSOs may be fairly consistent over time (around 106 MPN/100 ml for fecal coliform).
If sampling yields consistent results over time, the permittee may find that additional bacteria sampling is not
informative. Concentration data could be combined with flow data to determine bacteria loadings.
l pH
l Settleable solids
l Nutrients
The monitoring and modeling plan should also include monitoring for any other pollutants
for which water quality criteria are being exceeded, as well as pollutants suspected to be present in
the combined sewage and those discharged in significant quantities by industrial users. For example,
if the water quality criterion for zinc is being exceeded in the receiving water, zinc should be
monitored in the portions of the CSS where industrial users discharge zinc to the collection system.
POTW monitoring data and industrial pretreatment program data on nondomestic discharges can
help identify other pollutants that should be monitored. In coastal systems, measurements of
sodium, chloride, total dissolved solids, or conductivity can be used to detect the presence of sea
water in the CSS, which may be the result of intrusion through failed tide gates.
Not all pollutants need to be analyzed for each location sampled. For example:
l A larger list of pollutants should be analyzed for an industrial area suspected to have
contaminated storm water or a large load of pollutants in its sanitary sewer.
l Bacteria should be analyzed in a CSO upstream of a beach or drinking water supply with
past bacteriological problems, while it may not be necessary to analyze for metals or
other toxics.
3
The permittee should consider sampling both dissolved and total recoverable metals. The dissolved portion is more
immediately bioavailable, but does not account for metals that are held in solids. Since CSOs generally contain elevated
levels of suspended solids, which can release metals over time, sampling for total metals is important for evaluating
CSOs and their impacts.
The permittee should also ensure that monitored parameters correspond to the downstream
problem as well as the water quality criteria that apply in the receiving water body at the discharge
pipe. For example, the downstream beach may have an Enterococcus standard while the water
quality criterion at the discharge point might be expressed in fecal coliforms. In this case, samples
should be analyzed for both parameters.
The permittee should consider collecting composite data for certain parameters on as many
overflows as possible during the monitoring program. This can help establish mean pollutant
concentrations for computing pollutant loads. For instance, TSS concentrations are generally
important both because of potential habitat impacts and because they are associated with adsorbed
toxics. Collecting some discrete TSS samples can also be useful, particularly for evaluating the
existence of first flush.
The permittee should consider initial screening-level sampling for a wide range of pollutants
if sufficient information is not available to initially identify the parameters of concern. The
permittee can then analyze subsequent samples only for the subset of pollutants identified in the
screening. However, because pollutant concentrations in CSO discharges are highly variable, the
permittee should exercise caution in removing pollutants from the analysis list.
4
The potential significance of storm water discharges can often be assessed by looking at land uses and the relative
sizes of discharges.
particularly solids, oil and grease, and trace metals. For guidance on characterizing and monitoring
urban runoff, permittees can refer to EPA’s NPDES Storm Water Sampling Guidance Document
(U.S. EPA, 1992) and the Guide for Collection, Analysis, and Use of Urban Storm Water Data
(Alley, 1977).
The monitoring and modeling plan should reflect storm water and other sampling programs
occurring concurrently and provide for coordination with them. This will ensure that wet weather
discharges and their impacts are monitored and addressed in a cost-effective, targeted manner. Many
communities operate their storm water programs under a different department or authority from their
sewer program. Whenever possible, similar activities within these different organizations should
be coordinated on a watershed basis.
l Define the baseline conditions in the receiving water (chemical, biological, and physical
parameters)
The monitoring program should also provide background data on conditions in the receiving
waters during dry weather conditions, if this information is not already available (see Chapter 3).
Dry weather monitoring of the receiving water body helps define the background water quality and
will determine whether water quality criteria are being met or exceeded during dry weather.
Where a permittee intends to eliminate CSOs entirely (i.e., separate its system), only limited
or short-term receiving water monitoring may be necessary (depending on how long elimination of
CSOs will take). It may be useful, however, to collect samples before separation to establish the
baseline as well as after separation to evaluate the impacts of CSO elimination.
The permittee should coordinate monitoring activities closely with the NPDES permitting
authority. In many cases, it may be appropriate to use a phased approach in which the receiving
water monitoring program focuses initially on determining the pollutant loads from CSOs and
identifying short-term water quality impacts. The information obtained from the first phase can then
be used to identify additional data and analytical needs in an efficient manner. Monitoring efforts
can be expanded as circumstances dictate to provide additional levels of detail, including evaluation
of downstream effects and longer term effects.
The scope of the receiving water monitoring program will depend on several factors, such
as the identity of the pollutants of concern, whether the receiving water will be modeled, and the
relative size of the CSO. For example:
l To study dissolved oxygen (DO) dynamics, depth and flow velocity data must be
collected well downstream of the CSO outfalls. DO modeling may require data on the
plant and algae community, the temperature, the sediment oxygen demand, and the
shading of the river. Therefore, DO monitoring locations would likely span a larger area
than for some other pollutants of concern.
l When the volume of the overflow is small relative to the receiving water body, as in the
5
case of a small CSO into a large, well mixed river, the overflow may have little impact.
Such a situation generally would not require extensive downstream sampling.
In developing the monitoring and modeling plan, the permittee should consider the location
and impacts of other sources of pollutant loadings. As mentioned in Chapter 3, information on these
sources is generally compiled and reviewed during the initial system characterization. To evaluate
the impacts of CSOs on the receiving water body, the permittee should try to select monitoring
locations that have limited or known effects from these other sources, If the initial system
5
In areas where the receiving water is used for swimming, the dilution needs to be at least 10,000 to 1 for bacteria.
characterization did not provide sufficient information to adequately determine the location of these
sources, the permittee may need to conduct some monitoring to better characterize them.
In planning where to sample, it is important to understand land uses in the drainage basin
(which affect what pollutants are likely to be present) and characteristics of the receiving water body
such as:
l Flow patterns
l Changing conditions at individual sampling stations before, during, and after storm
events
In selecting monitoring locations, the permittee needs to consider physical logistics (e.g.,
whether the water is navigable, if bridges are available from which to sample) and crew safety.
Exhibit 4-2 illustrates how sampling locations might be distributed in a watershed to assess
the effect of other sources of pollution. If monitoring is conducted at the potential sampling
locations (labeled 1-6 in Exhibit 4-2), the results from the different locations could be compared to
provide a relative measure of the pollutant contributions from each source.
The permittee should also consider making cooperative sampling arrangements when
pollutants from multiple sources enter a receiving water or when several agencies share the cost of
the collection system and the POTW. The identification of new monitoring locations should account
for sites that may already be part of an existing monitoring system used by local or State government
agencies or research organizations.
In general, the monitoring and modeling plan should target receiving water monitoring to
those seasons, flow regimes, and other critical conditions where CSOs have the greatest potential
for impacts, as identified in an initial system characterization (see Chapter 3). It should specify
additional monitoring as necessary to fill data gaps and to support receiving water modeling and
analysis (see Tables B-2 through B-5 in Appendix B for potential modeling parameters), or to
determine the relative contribution of other sources to water quality impairment.
In establishing the frequency, duration, and timing of receiving water monitoring in the
monitoring and modeling plan, the permittee should consider seasonal variations to determine
whether measurable and significant changes occur in the receiving water body and uses during
different times of year. The monitoring and modeling plan should also enable the permittee to
address issues regarding attainment of WQS, such as:
l Assessing attainment of WQS for nutrients: This may call for samples collected
throughout the water body and timed to examine long-term average conditions over the
growing season.
l Assessing attainment of WQS for aquatic life support: This may call for biological
assessment in potentially affected locations and a comparison of the data to reference
sites.
l Reference site samples collected at separate locations for comparison with the CSO
study site to determine relative changes between the locations.
l Near-field studies to sample and assess receiving waters within the immediate mixing
zone of CSOs. These studies can examine possible short-term toxicity impacts or long-
term habitat alterations near the CSO.
l Far-field studies to sample and assess receiving waters outside the immediate vicinity
of the CSO. These studies typically examine delayed impacts, including oxygen
demand, nutrient-induced eutrophication, and changes in macroinvertebrate assemblages.
Section 4.6 discusses a strategy for determining whether to initiate monitoring for a particular wet
weather event.
The monitoring and modeling plan should identify parameters of concern in the receiving
water, including pollutants with water quality criteria for the designated use(s) of the receiving
water. The NPDES authority may have specific requirements or guidance regarding parameters for
CSO-related receiving water monitoring. These parameters may include the ones previously
identified for combined sewage (see Section 4.3.3):
l Indicator bacteria
l TSS
l BOD and DO
l pH
l Settleable solids
l Nutrients
l Metals (dissolved and total recoverable) and other toxics.
In addition, the permittee should consider the following types of monitoring prior to or concurrently
with the other analyses:
l Flow monitoring
l Biological assessment (including habitat assessment)
l Sediment monitoring (including metals and other toxics)
l Monitoring other pollutants known or expected to be present.
Monitoring should focus on the parameters of concern. In many cases, the principal concern
will be pathogens, represented by fecal coliform.
Depending on the complexity of the receiving water and the analyses to be performed, the
monitoring and modeling plan may need to reflect a larger list of parameters. Measuring
temperature, flow, depth, and velocity, and more complex parameters such as solar radiation, light
extinction, and sediment oxygen demand, can enable investigators to simulate the dynamics of the
6
receiving water that affect basic parameters such as bacteria, BOD, and TSS. Table B-1 in
Appendix B lists the data needed to perform the calculations for several dissolved oxygen, ammonia,
and algal studies. Indirect indicators, such as beach closings, fish advisories, stream bank erosion,
and the appearance of floatables, may also provide a relative measure of the impacts of CSOs.
The monitoring program should include enough storm events to enable the permittee to
predict the CSS’s response to rainfall events, the characteristics of resulting CSOs, and the extent
of impacts on receiving waters (as discussed in Sections 4.2.1, 4.3.2, and 4.5.2). By developing a
strategy for determining which storm events are most appropriate for wet weather monitoring, the
permittee can collect the needed data while limiting the number of times the sampling crew is
mobilized and the number of sampling events. This can result in significant savings in personnel,
equipment, and laboratory costs.
The following list (ORSANCO, 1998) contains key elements to consider in determining
whether to initiate monitoring for a wet weather event:
6
For example, a Streeter-Phelps DO analysis requires temperature, flow rate, reach length, and sediment oxygen
demand.
- Identify time periods contained within the monitoring schedule that may not be
representative of the system (holiday weekends) and avoid monitoring during those
periods
- Airports
- Municipalities
l Storm tracking
The flowchart in Exhibit 4-3 provides an example of how to apply these elements
(ORSANCO, 1998).
Exhibit 4-3. Decision Flowchart for Initiating a Wet Weather Monitoring Event
* Since the amount and intensity of precipitation needed to initiate an overflow and the physical conditions may vary significantly from CSO to CSO,
the permittee may need to establish different monitoring criteria for different CSOs.
The case study in Example 4-2 outlines the monitoring aspects of a comprehensive effort to
determine CSO impacts on a river and evaluate possible control alternatives. The city of South
Bend, Indiana developed and implemented a monitoring program to characterize flows and pollutant
loads in the CSOs and receiving water. The city then used a model to evaluate possible control
alternatives.
In developing its monitoring plan, South Bend carefully selected monitoring locations that
included roughly 74 percent of the area within the CSS and represented the most characteristic land
uses. The city conducted its complete monitoring program at 6 of the 42 CSO outfalls and
performed simpler chalking measurements at the remaining outfalls to give some basic information
on the occurrence of CSOs across the system. By using existing flow monitoring stations in the
CSS, the city was able to limit the need to establish new monitoring stations.
Since inaccurate or unreliable data may lead to faulty decisions in evaluating, selecting, and
implementing CSO controls, the monitoring and modeling plan must provide for quality assurance
and quality control to ensure that the data collected have the required precision and accuracy.
Quality assurance and quality control (QA/QC) procedures are necessary both in the field (during
sampling) and in the laboratory to ensure that data collected in environmental monitoring programs
are of known quality, useful, and reliable. The implementation of a vigorous QA/QC program can
also reduce monitoring expenses. For example, a QA/QC program for flow monitoring may help
prevent the need for resampling due to meter fouling or loss of calibration.
The City of South Bend, population of 109,000, has 42 combined sewer service areas covering over 14,000
acres.
Monitoring Goals
The ultimate goal of the CSO control effort was to reduce or eliminate impacts on uses of the receiving
water, the St. Joseph River. The more immediate goal consisted of quantifying CSO impacts to the St.
Joseph River and evaluating alternatives for cost-effective CSO control. To achieve these goals, the City
reviewed its existing data to determine what additionaldata were needed to characterize CSO impacts. The
City then developed and implemented a sampling and flow monitoring plan to fill in these data gaps.
Objectives of the monitoring plan included quantifying overflow volumes and pollutant loads in the
overflows and flows and pollutant loads in the receiving water. After evaluating various analytical and
modeling tools, the City decided to use the SWMM model to assist in predicting the benefits of alternative
control strategies and defining problems caused by CSOs.
The monitoring plan was designed to focus on the 6 largest drainage areas, which were most characteristic
of land uses within the CSS area and included 74 percent of that area, Monitoring all 42 outfalls was
judged to be unnecessarily costly. The monitoring plan specified 8 temporary and 9 permanent flow
monitoring locations along the main interceptor and in the influent and outfall structures of the 6 largest
CSOs. The interior surface of each non-monitored CSO diversion structure was chalked to determine
which storms caused overflows; after each storm, the depth to which the chalk disappeared was recorded.
Although the plan included monitoring only 14 percent of the outfalls, it measured flow and water quality
for most of the CSS area and covered a representative range of land uses and basins. Flow monitoring data
were used to calibrate the SWMM model.
The monitoring plan described water quality sampling procedures for both dry weather and wet weather
periods. The plan specified sample collection from four CSO structures during at least five storm events
representing a range of storm sizes. For the CSOs, monitored water quality parameters included nine
metals, total suspended solids (TSS), BOD, CBOD (carbonaceous biochemical oxygen demand), total
Kjeldahl nitrogen (TKN), ammonia, total phosphorus, total and fecal coliform bacteria, conductivity, and
hardness. Periodic dry-weather grab sample collections at the interceptors were also planned.
During storm events, water quality samples were collected using 24-bottle automatic samplers at the four
CSO points. To quantify “fist-flush” concentrations, the automatic samplers began collecting samples at
the start of an overflow event and continued collecting samples every five minutes for the first two hours
of the monitored events. A two-person crew drove between sites during each monitored event to check
equipment operation and the adequacy of sample collection.
River samples were taken from eight bridges along the St. Joseph River during and after three storms. Six
bridges are located within South Bend, and two are located just downstream in Michigan. River samples
were analyzed to determine the impacts of CSOs on the St. Joseph River and to calibrate and verify the
river model for dissolved oxygen, E. coli, and fecal coliform.
River samples were collected concurrently from the eight bridges every four hours. Four people sampled the
eight bridges. One person collected samples from two adjacent bridges within 30 minutes, Samples were
collected at the center of each bridge at the same location where the City collects its monthly river samples.
At least two sets of samples were collected before the storm to establish the baseline condition and the river
was sampled for at least 48 hours after onset of the storm to allow the river to return to its baseline condition.
Hourly rainfall data were collected from a network of five rain gages located in the drainage basins.
Results from the sampling and monitoring program for three storms during summer and early fall of 1991
indicated little or no impact on dissolved oxygen in the St. Joseph River, Large pulses in river bacteria counts
(E. coli and fecal coliform) were observed during the storms. Bacteria counts returned to baseline values
within 48 hours after the onset of each storm, Wet weather CSO sampling results showed a “first flush”
effect in three of the four sampled CSO structures. The fourth structure did not exhibit a “first flush” effect,
probably because of a high biochemical oxygen demand (BOD) loading at the upstream end of the trunk
sewer to the structure. Wet weather CSO sampling results also showed that the soluble metal concentrations
were much lower than the particulate metal concentrations.
The objective of the CSO control program is to solve real pollution problems and improve the river water
quality for specific uses. Based on the results of the monitoring program, bacteria reduction in the river
during wet weather has been the primary focus; A cost-performance curve was developed, using bacteria
reduction as the performance measure; to select the most cost-effective alternative and level of CSO control.
For an additional case study on CSO and receiving water monitoring, see Chapter 2 of Combined Sewer
Overflows - Guidance for Long-Term Control Plan (EPA, 1995a).
Quality assurance refers to programmatic efforts to ensure the quality of monitoring and
measurement data. QA programs increase confidence in the validity of the reported analytical data.
Quality control, which is a subset of quality assurance, refers to the application of procedures
designed to obtain prescribed standards of performance in monitoring and measurement. For
QC.
4-36
Chapter 4 Monitoring and Modeling Plan
Field QA/QC. QA programs for sampling equipment and for field measurement procedures
(for such parameters as temperature, dissolved oxygen, and pH) are necessary to ensure data are of
the appropriate quality. A field QA program should contain the following documented elements:
l
The sampling and analytical method; special sample handling procedures; and the
precision, accuracy, and detection limits of all analytical methods used.
l
The basis for selection of sampling and analytical methods. Where methods do not exist,
the QA plan should state how the new method will be documented, justified, and
approved for use.
l
Procedures for calibration and maintenance of field instruments and automatic samplers
during both dry and wet weather flows.
l
Training of all personnel involved in any function affecting data quality.
l
A performance evaluation system assessing the performance of field sampling personnel
in the following areas:
- Flow measurement
- The use of QC samples such as duplicate, split, or spiked samples and blanks as
appropriate to assess the validity of data.
l Procedures for recording, processing, and reporting data; procedures for use of non-
detects/results-below-detection in averaging or other statistical summaries (e.g.,
substituting one-half the detection level for results of non-detect at the lowest standard
used); procedures for review of data and invalidation of data based upon QC results.
l The amount of analyses for QC, expressed as a percentage of overall analyses, to assess
the validity of data.
l Equipment Blank - An aliquot of distilled water which is taken to and opened in the
field, its contents poured over or through the sample collection device, collected in a
sample container, and returned to the laboratory for analysis to check sampling device
cleanliness.
The permittee should also consider analyzing a sample of blank water to ensure that the water
is free of contaminants.
independently analyzed more than once, using the same methods and set of conditions. The
precision is estimated by the variability between repeated measurements. Accuracy refers to the
degree of difference between observed values and known or true values. The accuracy of a method
may be determined by analyzing samples to which known amounts of reference standards have been
added.
The following techniques are useful in determining confidence in the validity of analytical
data:
l Duplicate Samples (Laboratory) - Samples received by the laboratory and divided into
two or more portions at the laboratory, with each portion then separately and identically
prepared and analyzed. These samples assess precision and evaluate sampling
techniques and equipment.
l Split Samples (Field) - Single samples split in the field and analyzed separately check
for variation in laboratory method or between laboratories. Samples can be split and
submitted to a single laboratory or to several laboratories.
l Reagent Blanks - Preserving and analyzing a quantity of laboratory blank water in the
same manner as environmental water samples can indicate contamination caused by
sampling and laboratory procedures.
QA/QC programs are discussed in greater detail in EPA Requirements for Quality Assurance Project
Plans for Environmental Data Operations (U.S. EPA, 1994d) and Industrial User Inspection And
Sampling Manual For POTWs (U.S. EPA 1994c).
Although a permittee may collect accurate and representative data through its monitoring
efforts and verify the reliability of the data through QA/QC procedures, these data are of limited
usefulness if they are not stored in an organized manner and analyzed properly. The permittee
should develop a data management program to provide ready access to data, prevent data loss,
prevent introduction of data errors, and facilitate data review and analysis. Even if a permittee
intends to use a “complex” model to evaluate the impacts of CSOs and proposed CSO control
alternatives, the model still requires appropriate data for input parameters, as a basis for assumptions
made in the modeling process, and for model calibration and verification. Thus, the permittee needs
to properly manage monitoring data and perform some review and analysis of the data regardless of
the analytical tools selected.
All monitoring data should be organized and stored in a form that allows for ready access.
Effective data management is necessary because the voluminous and diverse nature of the data, and
the variety of individuals who can be involved in collecting, recording and entering data, can easily
lead to data loss or error and severely damage the quality of monitoring programs.
Data management systems must address both managerial and technical issues. The
managerial issues include data storage, data validation and verification, and data access. First, the
permittee should determine if a computerized data management system will be used. The permittee
should consider factors such as the volume of monitoring data (number of sampling stations, samples
taken at each station, and pollutant parameters), complexity of data analysis, resources available
(personnel, computer equipment, and software), and whether modeling will be performed. To enable
efficient and accurate data analysis, a computerized system may be necessary for effective data
management in all but the smallest watersheds. Computerized data management systems may also
facilitate modeling if the data can be uploaded directly into the model rather than being reentered.
Thus, when modeling will be performed, the permittee should consider compatibility with the model
when selecting any computerized data management system. Technical issues related to data
management systems involve the selection of appropriate computer equipment and software and the
design of the data system, including data definition, data standardization, and a data dictionary.
Data quality must be rigidly controlled from the point of collection to the point of entry into
the data management system. Field and laboratory personnel must carefully enter data into proper
spaces on data sheets and avoid transposing numbers. To avoid transcription errors when using a
computerized data management system, entries into a preliminary data base should be made from
original data sheets or photocopies. As a preliminary screen for data quality, the data
base/spreadsheet design should include automatic range-checking of all parameters, where values
outside defined ranges are flagged and either immediately corrected or included in a follow-up
review. For some parameters, it might be appropriate to include automatic checks to disallow
duplicate values. Preliminary data base/spreadsheet files should be printed and verified against the
original data to identify errors.
Additional data validation can include expert review of the verified data to identify possible
suspicious values. In some cases, consultation with the individuals responsible for collecting or
entering original data may be necessary to resolve problems. After all data are verified and
validated, they can be merged into the monitoring program’s master data files. For computerized
systems, to prevent loss of data from computer failure at least one set of duplicate (backup) data files
should be maintained.
During development of the monitoring and modeling plan, the permittee needs to consider
implementation issues such as recordkeeping and reporting requirements, personnel responsible for
carrying out each element of the plan, scheduling, and resources. Although some implementation
issues cannot be fully addressed in the monitoring and modeling plan until other plan elements have
evolved, they should be considered on a preliminary basis in order to ensure that the resulting plan
will satisfy reporting requirements and be feasible with available resources.
The monitoring and modeling plan includes a recordkeeping and reporting plan, since future
permits will contain recordkeeping and reporting requirements such as progress reports on NMC and
LTCP implementation and submittal of monitoring and modeling results. The recordkeeping and
reporting plan addresses the post-compliance monitoring program the permittee will develop as part
of the LTCP.
The monitoring and modeling plan identities the personnel that will implement the plan. In
some cases, particularly in a city with a small CSS, the appropriately trained personnel available for
performing the tasks specified in the monitoring and modeling plan may be very limited. By
reviewing personnel and assigning tasks, the permittee will be prepared to develop an
implementation schedule that will be attainable and will be able to identify resource limitations and
needs (including training) early in the process.
4.9.3 Scheduling
The monitoring and modeling plan has a tentative implementation schedule to ensure that
elements of the plan are implemented continuously and efficiently. The schedule can be revised as
necessary to reflect the review team’s assessment of the plan and the evaluation of monitoring and
modeling results. The schedule should address:
4.9.4 Resources
The monitoring and modeling plan identifies equipment, personnel, and other resource needs.
If modeling will be conducted, resource needs include a copy of the model and the equipment and
technical expertise to use the model. The plan may need to be modified after assessing the
availability of these resources. For example, if the monitoring and modeling plan identifies complex
modeling strategies, resource limitations may require the permittee to consider modeling techniques
that have more moderate data requirements. Alternatively, if the permittee does not have the
resources to purchase the hardware or software needed to run a detailed model, the permittee may
be able to make arrangements to use the equipment at another facility (e.g., another municipality
developing a CSO control program) or at a State or Federal agency. However, if such arrangements
are not possible, the permittee may need to choose a less detailed model which could lead to reduced
monitoring costs.
Through a review of resources, the permittee may identify monitoring equipment needed to
implement the monitoring and modeling plan. By obtaining needed equipment such as automatic
samplers, flow measuring equipment, rain gages, and safety equipment before the date when
monitoring is scheduled to begin, the permittee can prevent some potential delays.
CSS MONITORING
This chapter describes how to monitor rainfall, combined sewer system (CSS) flow, and
CSS water quality, and describes procedures for organizing and analyzing the data collected. It
discusses a range of monitoring and analysis options and provides criteria for identifying
appropriate options.
The CSO Control Policy identifies several possible objectives of a CSS monitoring program,
including:
l To adequately characterize the system’s response to wet weather events, such as the
volume, frequency, and duration of CSOs and the concentration and mass of pollutants
discharged
CSS monitoring also directly supports implementation of the following elements of the nine
minimum controls (NMC):
CSS monitoring will also support the in-depth system characterization and post-construction
compliance monitoring that are central elements in the LTCP.
This chapter outlines the steps that are critical to collection and analysis of rainfall, flow,
and water quality data in accordance with the CSO Control Policy.
Rainfall data are a vital part of a CSS monitoring program. This information is necessary
to analyze the CSS, calibrate and validate CSO models, and develop design conditions for
predicting current and future CSOs. Rainfall data should include long-term rainfall records and data
gathered at specific sites throughout the CSS.
This section describes how to install and use rainfall monitoring equipment and how to
analyze the data gathered.
The permittee’s rainfall data will probably include both national and local data. National
rainfall data are available from a number of Federal and local sources, including the National
Weather Service, the National Climatic Data Center (NCDC), airports, and universities (see
Chapter 3). Because rainfall conditions vary over short distances, the permittee will probably need
to supplement national data with data from local rainfall monitoring stations. Wastewater treatment
plants may already collect and maintain local rainfall data. If sufficient local rainfall data are not
available, the permittee may need to install rain gages. Where possible, the permittee should place
gages in every monitored CSO basin because of the high spatial variability of rainfall.
Equipment
Two types of gages are used to measure the amount and intensity of rainfall. A standard
rain gage collects the rainfall directly in a marked container and the amount of rain is measured
visually. Although inexpensive, standard gages do not provide a way to record changes in storm
intensity unless frequent observations are made during the storm.
Because wet weather flows vary with rainfall intensity, CSS monitoring programs typically
use recording gages, which provide a permanent record of the rainfall amount over time. The three
most common types of recording gages are:
l Weighing Type Gage - Water is weighed when it falls into a bucket placed on the
platform of a spring or lever balance. The weight of the contents is recorded on a chart,
showing the accumulation of precipitation.
l Float Recording Gage - Rainfall is measured by the rise of a float that is placed in the
collector.
l Gages should be located in open spaces away from the immediate shielding effects of
trees or buildings.
l Police, fire, public works, or other public buildings are desirable installation sites.
The permittee should synchronize rainfall monitoring with CSS flow monitoring, so that
rainfall characteristics can be related to the amount of runoff and CSO volume and a CSS model
can be calibrated and validated. In addition, long-term rainfall data gathered from existing gages
are necessary to develop appropriate design conditions for determining existing and future CSO
impacts on receiving water bodies. Because precipitation can vary considerably within short
distances, it is usually necessary to use data from several rain gages to estimate the average
precipitation for an area.
l Extend over a sufficient period of time (30 or more years is preferable; 10 is usually
acceptable); and
l Were collected close enough to the CSS’s service area to reflect conditions within that
area.
Common methods for characterizing rainfall include total volumes, event statistics, return
period/volume curves, and intensity-duration-frequency curves. These are described below.
Total Volumes. The National Weather Service publishes annual, monthly, and daily rainfall
totals, as well as averages and deviations from the average, for each rain gage in its network. The
time period for detailed simulation modeling can be selected by:
l Identifying wet- and dry-year rainfalls by comparing a particular year’s rainfall to the
long-term average; and
Simple hydraulic models can be used to predict total volumes of runoff, which can be used
to identify typical rainfall years and the variations across years. For example, 38 years of rainfall
records, 1955-1992, were collected at a NOAA gage near (but not within) a CSS drainage area.
These records indicate an average of 44 storm events per year, with a wide variation from year to
year. To generate runoff predictions for the CSS drainage area, the STORM runoff model (HEC,
1977) was calibrated and run using the 38 years of hourly rainfall data. The model predicted the
number of runoff events per year, the total annual runoff, and the average overflow volume per
event in inches/land area. Exhibit 5-1 ranks the years based on the number of events, inches of
runoff, and average runoff per event predicted by the model. Results showed the year 1969 had
both the highest number of runoff events (68) and largest total runoff volume (15.1 inches). The
year 1967 had the highest predicted average overflow per event (0.33 inches).
Exhibit 5-2 lists minimum, maximum, mean, and median values for the modeled runoff
predictions based on the data in Exhibit 5-1 for the example site. These statistics identify typical
and extreme years to select for modeling or predicting the frequency of overflows under various
control alternatives. Long-term computer simulations of the CSS using a multi-year continuous
rainfall record, or one-year simulations using typical or wet years, are useful for assessing
alternative long-term control strategies.
The data generated by the STORM model can be reviewed for typical or extreme years to
determine the uniformity of the monthly distribution of runoff. The years 1969 and 1956 represent
extreme high flows. The year 1956 had the most severe event over the 38-year evaluation period,
with 6.0 inches of runoff in 30 hours. The years 1970 and 1985 were selected as typical years,
having the most uniform distribution of rainfall throughout the year.
For some systems, the permittee may be able to identify typical years and analyze variations
by reviewing the rainfall record manually. In these cases, it may not be necessary to use a simple
hydraulic model to analyze rainfall data.
Exhibit 5-l. Ranking of Yearly Runoff Characteristics as Simulated by the Storm Model
January 1999
Chapter 5 CSS Monitoring
Exhibit 5-2. Rainfall and Runoff Parameters for Typical and Extreme Years
Return Period/Volume Curves. The “return period” is the frequency of occurrence for a
parameter (such as rainfall volume) of a given magnitude. The return period for a storm with a
specific rainfall volume may be plotted as a probability distribution indicating the percent of storms
with a total volume less than or equal to a given volume. For example, if approximately ten percent
of the storm events historically deposit 1.5 inches of rain or more, and there are an average of 60
storm events per year, an average of 6 storm events per year would have a total volume of 1.5 inches
or more, and the 1.5-inch rain event could be characterized as the “two-month storm.” Return
periods are discussed in Hydrology and Floodplain Analysis (Bedient and Huber, 1992).
Local rain gage data can be used to assess the applicability of the long-term record of the
site. For example, Exhibit 5-3 presents six weeks of local rainfall data from three tipping bucket
gages (labeled A, B, and C in Exhibit 5-4). Comparison with regional rainfall records indicates that
the average value of the three gages was close to the regional record with only slight variations
among gages.
Exhibit 5-3. 1993 Rainfall Data for a 5,305 Acre Drainage Area
Storm events 2, 4, and 8 were selected for detailed water quality sampling and analysis.
Subsequent analyses of CSS flow and CSS water quality data for this example are discussed in
Sections 5.3.3 and 5.4.2, respectively.
In cases where local rain gages are placed near but not exactly at the locations where CSS
flow and quality is being monitored, rainfall data from several nearby rain gage locations can be
interpolated to estimate the rainfall at the sampling location. The inverse distance weighting
method (see box on next page) can be used to calculate the rainfall over a CSS sampling location
in watershed 4 in Exhibit 5-4.
It may also be possible to use radar imaging data to estimate rainfall intensities at multiple
locations throughout the rainfall event.
January 1999
Chapter 5 CSS Monitoring
Using this method, the estimated precipitation at the sampling location is calculated as the weighted
average of the precipitation at the surrounding rain gages. The weights are the reciprocals of the squares of
the distances between the sampling location and the rain gages. The estimated rainfall at the sampling
location is calculated by summing the precipitation times the weight for each rain gage and dividing by the
sum of the u-eights. For example, if the distance between the sampling location in watershed 4 and ram
gage A is X, rain gage B is Y, and rain gage C is Z and the precipitation at each rain gage is PA, PB, and PC,
then the precipitation at the sampling location in watershed 4 can be estimated by:
Not to Scale
Chapter 5 CSS Monitoring
Flow measurement techniques vary greatly in complexity, expense, and accuracy. This
section describes a range of manual and automated flow monitoring techniques. Exhibit 5-5
summarizes their advantages and disadvantages.
Manual Methods
The simplest flow monitoring techniques include manual measurement of velocity and depth,
use of bottle boards and chalking (see Example 5-1), and dye testing. Manual methods are difficult
during wet weather, however, since they rely extensively on labor-intensive field efforts during
storm events and do not provide an accurate, continuous flow record. Manual methods are most
useful for instantaneous flow measurement, calibration of other flow measurements, and flow
measurements in small systems. They are difficult to use for measuring rapidly changing flows
because numerous instantaneous measurements must be taken at the proper position to correctly
estimate the total flow.
Velocity Meters
Ultrasonic Meter designed to measure velocity l Instrument does not interfere l More expensive than
through a continuous pulse with flow other equipment
l Can be used in full pipes
Electromagnetic Meter designed to measure velocity l Instrument does not interfere l More expensive than
through an electromagnetic process with flow other equipment
l Can be used in full pipes
flow rate calculation. Monitoring devices need to be resistant to fouling and clogging because of
the large amounts of grit and debris in a CSS.
Depth-sensing devices can be used with pipe equations or primary flow and velocity-sensing
devices to determine flow rates. They include:
l Ultrasonic Sensors, which are typically mounted above the flow in a pipe or open
channel and send an ultrasonic signal toward the flow. Depth computations are based
on the time the reflected signal takes to return to the sensor. These sensors provide
accurate depth measurements but can be affected by high suspended solid loads or
foaming on the water surface.
l Pressure Sensors, which use transducers to sense the pressure of the water above them.
They are used with a flow monitor that converts the pressure value to a depth
measurement.
l Bubbler Sensors, which emit a continuous stream of fine bubbles. A pressure transducer
senses resistance to bubble formation, converting it to a depth value. These devices
provide accurate measurements. The bubble tube can clog, however, and the device
itself requires frequent calibration.
l Float Sensors, which sense depth using a mechanical float, often within a chamber
designed to damp out surface waves. Floats can clog with grease and solid materials and
are, therefore, not commonly used to sense flow in sewers.
A bottle rack is used to determine the approximate depth of overflows from a 36-inch combined sewer in an
overflow manhole (Exhibit 5-6). The overflow weir for this outfall is 12 inches above the invert of the sewer,
and flows below this level are routed out the bottom of the structure to the interceptor and the wastewater
treatment plant. Any flow overflowing the 12-inch weir is routed to the 42-inch outfall sewer. Attached to
the manhole steps, the bottle rack approximates the flow level in the manhole by the height of the bottles that
are filled. This outfall has potential for surcharging because of flow restrictions leading to the interceptor.
Consequently, the bottle rack extends well above the crown of the outfall sewer. After each rainfall, a
member of the monitoring team pulls the rack from the manhole, records the highest bottle filled, and returns
the rack to the manhole, Exhibit 5-7 presents depth data for the nine storms listed in Exhibit 5-3.
Storm 3, which had 0.1 inch of rain in 85 minutes, was contained at the outfall with no overflow, although
it did overflow at other locations. Storm 5, with an average volume of 0.14 inches and an average intensity
of 0.09 in/hour, had a peak flow depth of approximately six inches above the weir crest.
It is instructive to examine the individual rain gages (located as indicated in Exhibit 5-4) and compare them
to the flow depths. Rain gage A indicated that Storms 3 and 5 had similar depths and that 3 was slightly more
intense. Why, then, did Storm 5 cause an overflow, while Storm 3 did not? Rain gage B, which lies nearer
to the outfall, indicates 50 percent more volume and 50 percent higher intensity for storm 5. Using only rain
gage A in calibrating a hydraulic model to the outfall for storms 3 and 5 could have posed a problem.
Because a bottle board indicates approximate maximum flow depth, not duration or flow volume, it is not
sufficient to calibrate most models.
Storms 4 and 8 caused flow depth to surcharge, or increase above the crown of the pipe. Both storms
occurred during late afternoon when sanitary sewer flows are typically highest, potentially exacerbating the
overflow, The surcharging pipe indicates that flow measurements will be difficult for large storms at this
location. Further field investigations will be necessary to define the hydraulics of this particular outfall and
intercepting device, Because of safety considerations in gaming access to this location, the monitoring team
used only the bottle board during the early monitoring period. Later, the team instailed a velocity meter and
a series of depth probes to determine a surface profile.
Section
January 1999
Chapter 5 CSS Monitoring
Using depth measurement data, pipe equations can be applied to develop flow estimates. The
Hazen-Williams equation, Manning equation, and similar equations can be useful for estimating flow
capacity of the system and performing a preliminary flow analysis of the CSS. The Hazen-Williams
equation is generally used for pressure conduits, while the Manning equation is usually used in open-
channel situations (Viessman, 1993). The Hazen-Williams equation is:
0.63 0.54
V = 1.318 C(R) (S)
where:
where:
Q=VA
where:
Since the calculations are based on the average upstream characteristics of the pipe, personnel
should measure depth at a point in the sewer where there are no bends, sudden changes in invert
elevation, or manholes immediately upstream. These features can introduce large errors into the
flow estimate. Anomalies in sewer slope, shape, or roughness also can cause large errors (50 percent
and greater) in flow measurement. However, in uniform pipes, a careful application of these
formulas can measure flows with an error as low as 10 to 20 percent (ISCO, 1989). The permittee
can improve the accuracy of the equation somewhat by calibrating it initially, using measurements
of velocity and depth to adjust slope and roughness values.
Velocity Meters
Velocity meters use ultrasonic or electromagnetic technology to sense flow velocity at a
point, or in a cross section of the flow. The velocity measurement is combined with a depth value
(from a depth sensor attached to the velocity meter) to compute flow volume. Velocity meters can
measure flows in a wider range of locations and flow regimes than depth-sensing devices used with
primary flow devices, and they are less prone to clogging. They are comparatively expensive,
however, and can be inaccurate at low flows and when suspended solid loads vary rapidly. One type
of meter combines an electromagnetic velocity sensor with a depth sensing pressure transducer in
a single probe. It is useful for CSO applications because it can sense flow in surcharging and
backflow conditions. This device is available as a portable model or for permanent installation.
Most flow monitoring involves the use of portable, battery-operated depth and velocity
sensors, which are left in place for several storm events and then moved elsewhere. For some
systems, particularly small CSSs, the monitoring program may involve manual methods. In such
cases, it is important to allocate the available personnel and prepare in advance for the wet weather
events.
Although temporary metering installations are designed to operate automatically, they are
subject to clogging in CSSs and should be checked as often as possible for debris.
Some systems use permanent flow monitoring installations to collect data continuously at
critical points. Permanent installations also can allow centralized control of transport system
facilities to maximize storage of wastewater in the system and maximize flow to the treatment plant.
The flow data recorded at the site may be recovered manually or telemetered to a central location.
To be of use in monitoring CSSs, flow metering installations should be able to measure all
possible flow situations, based on local conditions. In a pipe with smooth flow characteristics, a
weir or flume in combination with a depth sensor or a calibrated Manning equation may be
sufficient. Difficult locations might warrant redundant metering and frequent calibration. The key
to successful monitoring is combining good design and judgment with field observations, the
appropriate metering technology, and a thorough meter maintenance and calibration schedule.
The CSS flow data can be evaluated to develop an understanding of the hydraulic response
of the system to wet weather events and to answer the following questions for the monitored outfalls:
l What size storm can be contained by the regulator serving each outfall? What rainfall
amount is needed to initiate overflow? Does this containment capacity vary from storm
to storm?
l Approximately how many overflows would occur and what would be their volume,
based on a rainfall record from a different year? How many occur per year, on average,
based on the long-term rainfall record?
Extrapolating from the monitored period to other periods, such as a rainfall record for a year
with more storms or larger volumes, requires professional judgment and familiarity with the data.
For example, as shown in Exhibit 5-8, the flow regulator serving Outfall 4 prevented overflows
during Storm 3, which had 0.10 inch of rain in 1.4 hours. However, approximately half of the
rainfall volume overflowed from Storm 5, which had 0.14 inch in 1.5 hours. From these data, the
investigator might conclude that, depending on the short-term intensity of the storm or the
antecedent moisture conditions, Outfall 4 would contain a future storm of 0.10 inches but that even
slightly larger storms would cause an overflow. Also, Exhibit 5-8 indicates that a storm even as
small as Storm 3 can cause overflows at the other outfalls.
V = overflow volume (inches depth when inches of overflow is spread over drainage area)
R = rainfall depth (inches)
na = no measurement available
Chapter 5 CSS Monitoring
Comparing the overflow volumes of different outfalls indicates which outfalls contribute the
bulk of the overflow volume and, depending on loading measurements, may contribute most heavily
to water quality problems. To compare the hydraulic performance of different outfalls, flows should
be normalized against the drainage area and rainfall. Provided that rainfall data are representative
of the area’s rainfall, inches of overflow (spread over the discharge subarea) per inch of rainfall
constitutes a useful statistic. Exhibit 5-8 presents the overflow volumes in inches and the ratio of
depth of overflow to depth of rain (V/R).
For each outfall, V/R varies with the storm depending on the number of antecedent dry days,
the time of the storm, and the maximum rainfall intensity. V/R also varies with the outfall
depending on land characteristics such as its impervious portion, the hydraulic capacity upstream
and downstream of the flow regulator, the operation of the flow regulator, and features that limit the
rate at which water can enter the system draining to that overflow point. Because of the large
number of factors affecting variations in V/R, small differences generally provide little information
about overflow patterns. However, certain patterns, such as an increase in V/R over time or large
differences in V/R between storms or between outfalls, may indicate design flaws, operational
problems, maintenance problems, or erroneous flow measurements, or a rainfall gage that does not
represent the average depth of rain falling on the discharge subarea.
In addition to supporting an analysis of CSO volume, flow data can be used to create a plot
of flow and head for a selected conduit during a storm event, as shown in Exhibit 5-9. These plots
can be used to illustrate the conditions under which overflows occur at a specific outfall. They can
also be used during CSS model calibration and verification (see Chapter 7).
Exhibits 5-8 and 5-9 (representing different CSS monitoring programs) illustrate some of the
numerous methods available for analyzing CSO flow monitoring data. Additional methods include
plotting regressions of overflow volume and rainfall to interpret monitoring data and identify
locations that will cause difficulty in calibrating a model. For this type of regression, the y-intercept
defines the rainfall needed to cause an overflow and the slope roughly defines the gross runoff
coefficient for the basin. Flow data can also be used to tabulate CSO volumes and frequencies
during the monitored time period and to compare the relative volumes and frequencies from different
monitoring sites in the CSS. Data are plotted, tabulated, and analyzed prior to a modeling
assessment (described in Chapter 7).
This section outlines various methods for collecting, organizing, and analyzing CSS wastewater data.
Sampling during wet weather events involves some factors that are not a significant concern during
dry weather. These additional considerations are discussed in the section on sample program
organization for receiving water quality monitoring (Section 6.3.1).
In general, wastewater sample types fall into the following two categories:
l Grab samples
l Composite samples.
Grab Sampling. A grab sample is a discrete, individual sample collected over a maximum
of 15 minutes. Grab samples represent the conditions at the time the sample is taken and do not
account for variations in quality throughout a storm event. Multiple grab samples can be gathered
at a station to define such variations, although costs increase due to additional labor and laboratory
expenses.
l Collect samples at equal time intervals at a volume proportional to the flow rate (e.g.,
collect 100 ml of sample for every 100 gallons of flow that passed during a lo-minute
interval).
l Collect samples of equal volume at varying times proportional to the flow (e.g., collect
a 100 ml sample for each 100 gallons of flow irrespective of time).
The second method is preferable for sampling wet weather flows, since it results in the
greatest number of samples when the flow rate is the highest. More detailed information on methods
of flow weighting is presented in the NPDES Storm Water Sampling Guidance Document (U.S.
EPA, 1992).
Grab and composite samples can be collected using either of two sample methods: manual
and automatic.
Manual Sampling. Manual samples are usually collected by an individual using a hand-held
container. This method requires minimal equipment and allows field personnel to record additional
observations while the sample is collected. Because of their special characteristics, certain pollutants
should be collected manually. For example, fecal streptococcus, fecal coliform, and chlorine have
very short holding times (i.e., 6 hours), pH and temperature need to be analyzed immediately, and
oil and grease can adhere to the sampling equipment and cause inaccurate measurements. Volatile
compounds must be collected manually according to standard procedures since these compounds will
likely volatilize as a result of agitation during automatic sampler collection (APHA, 1992).
Manual sampling can be labor-intensive and expensive when the sampling program is long-
term and involves many locations. Personnel must be available around the clock to sample storm
events. Safety issues or hazardous conditions may affect sampling at certain locations.
Automated Sampling. Automated samplers are useful for CSS sampling because they can
be programmed to collect multiple discrete samples as well as single or multiple composited
samples. They can collect samples on a timed basis or in proportion to flow measurement signals
from a flow meter. Although automated samplers require a large investment, they can reduce the
amount of labor required in a sampling program and increase the reliability of flow-weighted
cornpositing.
Automated samplers have a lower compartment, which holds glass or plastic sample
containers and an ice well to cool samples, and an upper part, containing a microprocessor-based
controller, a pump assembly, and a filling mechanism. The samplers can operate off of a battery,
power pack, or electrical supply. More expensive samplers have refrigeration equipment and require
a 120-volt power supply. Many samplers can be connected to flow meters that will activate flow-
weighted cornpositing programs, and some samplers are activated by inputs from rain gages.
l Some pollutants (e.g., oil and grease) cannot be sampled by automated equipment unless
only approximate results are desired.
l The self-cleaning capability of most samplers provides reasonably separate samples, but
some cross-contamination is unavoidable because water droplets usually remain in the
tubing.
l Debris in the sewer, such as rags and plastic bags, can block the end of the sampling line,
preventing sample collection. When the sampling line is located near a flow meter, this
clogging can also cause erroneous flow measurements. Samplers and meters should be
checked during storms and must be tested and serviced regularly. If no field checks are
made during a storm event, data for the entire event may be lost.
l The sample nozzles of many automatic samplers do not have the velocity capabilities
necessary for picking up the sand and gravel in untreated CSO flows.
Sampling Strategies
In developing a sampling strategy, the permittee should consider the timing of samples and
sampling intervals (i.e., duration of time between the collection of samples). Since pollutant
concentrations can vary widely during a storm event, the permittee should consider sampling
strategies that include pre-storm, first flush, peak flow, recovery, and post-storm samples. For
example, the permittee could take individual grab samples at each site during the different storm
stages. Another sampling regime the permittee can use is taking a series of samples during the
stages at each site:
A third possible sampling regime could include a first flush composite taken over the first
30 minutes of discharge, followed by a second composite over the next hour of discharge, followed
by a third composite for the remainder of the storm. To characterize first flush, a sample should be
collected as close to the beginning of the CSO event as feasible. Appropriate sampling intervals
depend on such factors as drainage area sizes, slopes, land uses, and percent imperviousness.
l Bacteria - Because samples collected for bacteria analysis cannot be held for more than
six hours, they must be collected manually. Bacteria samples are collected directly into
a sterile container or plastic bag, and it is important not to contaminate the sample by
touching it. Often the samples are preserved with sodium thiosulfate.
l Volatile Organic Compounds (VOCs) - Samples analyzed for VOCs are collected
directly into special glass vials. Each vial must be filled so that there is no air space into
which the VOCs can volatilize and be lost.
l Oil and Grease - Samples analyzed for oil and grease must be collected by grab sample
using a glass jar with a Teflon-coated lid. Samples are preserved by lowering the pH
below 2.0 using a strong acid.
l Dissolved Metals - Samples collected for dissolved metals analysis must be filtered
immediately after sample collection and before preservation.
The monitoring program may also include toxicity testing, in which the acute and chronic
impacts to aquatic life are determined. Toxicity testing procedures for wet weather discharges are
in Technical Support Document for Water Quality-based Toxics Control (U.S. EPA, 1991a).
Water samplers, sampling hoses, and sample storage bottles should always be made of
materials compatible with the pollutants being sampled. For example, when sampling for metals,
bottles should not have metal components that can contaminate the samples. Similarly, bottles and
caps used for organic samples should be made of materials not likely to leach into the sample.
Name of project
Date and time of sample collection
Sample location
Name or initials of sampler
Analysis to be performed
Sample ID number
Preservative used
Type of sample (grab, composite).
Regulations (49 CFR Parts 171 - 177). Air shipment of samples classified as hazardous may also be
covered by the Dangerous Goods Regulations (International Air Transport Association, 1996).
Samples should be sealed with chain-of-custody form seals in leak-proof bags and padded
against jarring and breakage. Samples must be packed with an ice substitute to maintain a
temperature of 4°C during shipment. Plastic or metal coolers make ideal shipping containers
because they protect and insulate the samples. Accompanying paperwork such as the chain-of-
custody documentation should be sealed in a waterproof bag in the shipping container.
The sampler and the laboratory should retain copies of the chain-of-custody form. Contract
laboratories often supply chain-of-custody forms with sample containers. The form is also useful
for documenting which analyses will be performed on the samples. Forms typically contain the
following information:
Since monitoring programs can generate large amounts of information, effective management
and analysis of the data are essential. Even small-scale programs, such as those involving only a few
CSS and receiving water monitoring locations, can generate an extensive amount of data. This
section discusses tools for data analysis including spreadsheets, graphical presentations, and
statistical analysis. (Data management is discussed in Section 4.8.2. Chapters 7 and 8 discuss more
detailed data analysis during modeling.)
This section outlines an example analysis of data collected during three storms, where flow-
weighted composite samples were collected and analyzed for BOD and TSS. Exhibit 5-10 shows
average concentrations for each storm at the monitored outfalls; the small sample size does not
provide statistically reliable information on the expected variability of these concentrations for other
events. To estimate pollutant concentrations for a large set of storm events, expected values can be
calculated by assuming a lognormal distribution. (The lognormal distribution has been shown to be
applicable to CSO quality (Driscoll, 1986).) Exhibit 5-11 shows that the mean and median for the
data are similar and are within typical ranges for CSO quality. The mean and median for the
sampling data can be used with a lognormal distribution to compute the expected mean, median, and
90th-percentile value for a large data set of many storm events. If used as a basis for estimating
impacts, the 90th-percentile values would be more conservative than the means for BOD and TSS
since only 10 percent of the actual concentrations for these pollutants should exceed the 90th-
percentile values.
Multiplying flow measurements (or estimates) by pollutant concentration values drawn from
monitoring data gives the total pollutant load discharged during each storm at each outfall.
Exhibit 5-12 lists pollutant loads for the three storms at each monitored outfall. As with flow data,
these brief statistical summaries provide insight into the response of the system before any more
involved computer modeling is performed. For example, the load in pounds of BOD and TSS
discharged at each outfall, normalized to account for differences in rainfall depth or land area at each
outfall, helps to identify differences in loading rates across outfalls over the long term. These
loading factors can provide rough estimates of the loads from unmonitored outfalls that have land
BOD TSS
Mean 96.87 345.27
Median 94.00 350.00
Expected Mean* 97.16 352.53
Expected Median* 94.70 321.29
Expected 90th Percentile Value* 126.64 558.03
1
Typical CSO Characteristics 60 - 220 270 - 550
*Projected statistic from sampling population (i.e., very large data set)
1
Metcalf & Eddy, Inc., 1991.
uses or impervious areas similar to the monitored area. Finally, the total load per storm helps in
comparing storms and projecting storm characteristics that would produce higher or lower loads.
Pollutant loads are affected by the number of dry days and the number of days without a flushing
storm because these factors represent a period when no severe scour activity occurred in the sewer
system.
Three storms can indicate trends but do not provide enough data to characterize the load of
the CSS or its individual source areas. As additional data are collected during the monitoring
program, estimates based on the data set become statistically more reliable because the size of the
data sets increases. The additional information allows continual refinement of the permittee’s
knowledge of the system.
The example shown in Exhibit 5-13, involving bacteria sampling, illustrates the value of
correlating flow and concentration data. Because automated samplers are not appropriate for
collecting bacterial samples, manual grab samples were collected and analyzed for fecal coliform
bacteria. During a single storm event, samples were collected from Outfall 1 at 30 minute intervals,
beginning shortly after the storm started and ending with sample #6 approximately 2½ hours later.
Peak flow occurred within the first 90 minutes. The fecal coliform concentration peaked in the first
half hour and declined nearly one-hundredfold to the last sample, exhibiting a “first flush” pattern.
6
The average concentration was 3.14 x 10 MPN/100 ml. To calculate total fecal coliform loading,
flow measurements were multiplied by the corresponding grab sample concentrations at each half-
hour interval, as shown in the right-hand column. The average concentration was also multiplied
14
by the total flow for comparative purposes. This second calculation (1.79 x 10 MPN)
overestimates the total loading, primarily because it fails to correlate the decreasing bacteria level
to the changing flows.
In many cases background conditions or upstream wet weather sources, such as separate
storm sewer systems, may provide significant pollutant loads. Where possible, the permittee should
try to assess loadings from non-CSO sources in order to fully characterize the receiving water
impacts from CSOs. In some cases, these other sources may be outside the permittee’s jurisdiction.
If the permittee cannot obtain existing monitoring data on these sources, the permittee should
consider monitoring these sources or entering into an agreement to have the appropriate party
conduct the monitoring. The data analysis techniques discussed in this section apply equally well
to other wet weather sources, although the pollutant concentrations in such sources may differ
significantly.
* For CSOs, fecal coliform concentrations typically range from 2.0 x 105 - 1.1 x 106 colonies/100 ml (Metcalf &
Eddy, 1991).
** Load = [Concentration (No./100 ml) x Total Flow (ml)] / 100 (since concentration is for 100 ml)
Total Flow (in ml) = cfs x 1800 (# of seconds in one 30-minute interval) x 28,321 (# of ml in one cf)
*** Load estimated by multiplying the average bacteria concentration by the total flow
Single composite samples or average data may be sufficient for a preliminary estimate of
pollutant loadings from CSOs. Establishing an upper-bound estimate for such loads may be
necessary in order to analyze short-term impacts based on short-term pollutant concentrations in the
receiving water and to develop estimates for rarer events that have not been measured. A statistical
distribution, such as normal or lognormal, can be developed for the data and mean values and
variations can be estimated. These concentrations can be multiplied by measured flows or an
assumed design flow to generate storm loads in order to predict rare or extreme impacts. Chapters 8
and 9 discusses further how to predict receiving water impacts.
The case study in Example 5-2 presents an approach for sampling and data analysis used by
1
Columbus, Georgia. The City found this approach useful in assessing CSO control options.
The City of Columbus, Georgia, in a CSO technology demonstration project, found significant correlation
between the timing and volume of CSO pollutant loadings and the pre-storm dry weather conditions. These
relationships can be used for:
The Approach
The approach involves conducting discrete sampling (for flow and water quality) and using these sampling
results and historical rainfall data to establish annual load and design rate relationships (% of annual quantity vs.
design flow for volume and pollutants). The discrete sampling is timed to obtain more samples at the beginning
of the storm event and fewer samples as the event progresses (pollutant weighted sampling}. Using this
sampling plan in Columbus has resulted in data that show a significant correlation between the cumulative
volume and pollutant mass for different pre-storm conditions.
Flow measurements can be correlated with rain rate measurements to establish a rainfall/runoff relationship for
the total event and rainfall intensity. These pollutant and runoff correlations are used with the historical rainfall
data to quantify annual pollutant loads and to define a relationship between design rate and annual quantity for
control or treatment.
These relationships can be used to evaluate any specific control or various combinations of controls and define
annual pollutant quantities for each control level. Types of controls include collection system maximization of
flows and attenuation, storage, and direct treatment.
The entire procedure can be applied using simple spreadsheet methods or can be incorporated into more
sophisticated modeling efforts.
The methodology can be used in either the presumption or demonstration approaches. In the presumption
approach, where the objective is to treat the mass from 85% of the annual volume using primary clarification,
the Columbus method can show that the objective can be reached with facilities at much smaller flow rates by
applying better treatment to the more polluted, more frequent rainfall events. The net result can be less costly to
facilities.
1
The specific approach used by Columbus, GA, may not be appropriate for all CSO communities.
Cost-benefit levels of control can be determined from “knee-of-the curve” analyses using design rate
relationships, and may represent different annual objectives for different pollutants to be reduced. For
example:
l Treatment rate versus percent annual pollutant treated can be used to define the design storm
criteria
l Treatment rate versus percent annual CSO volume treated can be used to define the level of high
rate disinfection.
Alternatively, different levels of control can be evaluated to estimate the end-of-pipe loads and resulting in-
stream concentrations for various flows. This provides a historical distribution of in-stream concentrations
that can be compared to a waste load allocation to define statistical exceedances in a wet weather permit.
Finally, the evaluated treatment options can be compared using life-cycle costs and pollutant removal results,
Fur chemical disinfection, the TSS loading relationship can be used in controlling the rate of disinfection.
The disinfectant feed is varied according to the variation of incoming solids to accomplish the disinfection
objective while minimizing the potential for overdosing.
This chapter discusses techniques and equipment for receiving water monitoring, including
hydraulic, water quality, sediment, and biological sampling procedures. The techniques vary in
applicability and complexity, but all are generally applicable to CSO-impacted receiving waters.
In collecting and analyzing receiving water monitoring data, the permittee needs to implement a
quality assurance and quality control (QA/QC) program to ensure that accurate and reliable data are
used for CSO planning decisions (see Section 4.8.1). For purposes of the post-construction
compliance monitoring program, all sampling and analysis needs to be done in accordance with EPA
regulations.
The CSO Control Policy discusses characterization and monitoring of receiving water
impacts as follows:
l In order to design a CS0 controlplan adequate to meet the requirements of the CWA,
a permittee should have a thorough understanding of its sewer system, the response of
the system to various precipitation events, the characteristics of the overflows, and the
water quality impacts that result from CSOs.
l The permittee should adequately characterize...the impacts of the CSOs on the receiving
waters and their designated uses. The permittee may need to consider information on
the contribution and importance of other pollution sources in order to develop a final
plan designed to meet water quality standards.
l
The permittee should develop a comprehensive, representative monitoring program that
. . . assesses the impact of the CSOs on the receiving waters. The monitoring program
should include necessary CSO effluent and ambient in-stream monitoring and, where
appropriate, other monitoring protocols such as biological assessment, toxicity testing
and sediment sampling. Monitoring parameters should include, for example, oxygen
demanding pollutants, nutrients, toxic pollutants, sediment contaminants, pathogens,
bacteriological indicators (e.g., Enterococcus, E. Coli), and toxicity. A representative
sample of overflow points can be selected that is sufficient to allow characterization of
CSO discharges and their water quality impacts and to facilitate evaluation of control
plan alternatives. (Section II.C.1)
As discussed in Chapter 2, the CSO Policy intends for the permittee to use either the
presumption approach or the demonstration approach in identifying controls that will provide for
attainment of water quality standards (WQS). Under the demonstration approach, the permittee
demonstrates the adequacy of its proposed CSO control program to attain WQS. Generally,
permittees selecting the demonstration approach will need to monitor receiving waters to show that
their control programs are adequate.
The presumption approach is so named because it is based on the presumption that WQS will
be attained when certain performance-based criteria identified in the CSO Policy are achieved, as
shown by the permittee in its long-term control plan (LTCP). The regulatory agency is likely to
request some validation of the presumption, such as receiving water quality sampling or end-of-pipe
sampling of overflows combined with flow information and dilution calculations. Chapters 7 (CSS
Modeling) and 8 (Receiving Water Modeling) discuss the different modeling considerations related
to the demonstration and presumption approaches.
When a CSO enters a receiving water body, it is subject to fate and transport processes that
modify pollutant concentrations in the receiving water body. The impact of CSOs on receiving
waters is largely determined by the hydraulics of the receiving water body and the relative magnitude
of the CSO loading. Assessing receiving water hydraulics is an important first step in a receiving
water study, since an understanding of how CSOs are transported and diluted is essential to
characterizing their impacts on receiving waters. Awareness of large-scale and small-scale
hydrodynamics can help the permittee determine where to sample in the receiving water for the
effects of CSOs. Large-scale water movement largely determines the overall transport and
transformation of pollutants. Small-scale hydraulics, such as water movement near a discharge point
(often called near-field), determine the initial dilution and mixing of the discharge, For example,
a discharge into a wide, fast-flowing river might not mix across the river for a long distance since
it will quickly be transported downstream.
Hydraulic monitoring involves measuring the depth and velocity of the receiving water body
and its other physical characteristics (e.g., elevation, bathymetry, cross section) in order to assess
transport and dilution characteristics. This may include temporary or permanent installation of gages
to determine depth and velocity variations during wet weather events. In all cases, the permittee will
need to use existing mapping information or perform a new survey of the physical characteristics
of the receiving water in order to interpret the hydraulic data and understand the hydraulic dynamics
of the receiving water. (Section 4.5 discusses receiving water sampling designs and the selection
of monitoring locations.)
Identifying a suitable hydraulic monitoring method depends largely on the type and
characteristics of receiving water.
In rivers and streams, flow rate is generally a factor of the depth, width, cross-sectional area,
and hydraulic geometry of the river or stream channel. Flow in rivers and streams is usually
determined by measuring the stage (elevation of water above a certain base level) and relating stage
to discharge with a rating curve. This relationship is developed by measuring flow velocity in the
stream or river at different stages, and using velocity and the area of the stream or river channel to
determine the total discharge for each stage (Bedient and Huber, 1992). For large rivers and streams,
long-term flow and geometry data are often available for specific gaging stations from USGS and
the U.S. Army Corps of Engineers.
For a CSO outfall located near a USGS gage, the monitoring team can use the nearest USGS
1
gage watershed areas to estimate flow at the discharge site. Flow information may also be available
from stage measurements at bridge crossings and dams, and from studies performed by other State
and Federal agencies. In the absence of such flow data, the permittee may need to install stage
indicators or use current meters to collect stream flow measurements. Some of the CSO flow
monitoring devices described in Exhibit 5-5 of Chapter 5 also may be used to measure open channel
flow in rivers and streams. The USGS (1982) and USDI (1984) have published detailed manuals
on stream gaging and measurement techniques.
Estuaries
Estuaries connect rivers and oceans and thus represent a complex system of tides, salinity,
and upstream drainage. Tidal variations and density effects from the varying levels of salinity need
to be defined to determine how pollutants from CSOs are transported.
Tidal variations affect estuarine circulation patterns which, along with salinity patterns,
determine how pollutant loadings entering the estuary are dispersed. Based on velocity and salinity
patterns, estuaries can be classified as one of the following types:
l Stratified estuaries have a large fresh water inflow over a more dense salt water layer.
Tidal currents are not sufficient to mix the separate layers. Transport of pollutants is
largely dependent on the difference in the densities of the pollutants and the receiving
water.
l Well-mixed estuaries have a tidal flow much greater than the river outflow, with mixing
and flow reversal sufficient to create a well-mixed water column at all depths. Pollutants
tend to move with the motion of the tides and are slowly carried seaward.
l Partially-mixed estuaries have flow and stratification characteristics between the other
two types and have tide-related flows much greater than river flows. Pollutant transport
depends somewhat on density, but also involves significant vertical mixing.
1
For example, the 5,000-square mile Merrimack River watershed in New Hampshire and Massachusetts
has 46 USGS gages that monitor most of the larger tributaries and the main stem in several locations. Using flow
data from the one or two gages closest to a CSO outfall, flow at the outfall can be estimated based on the relative
watershed areas between the gages and the outfall.
Classification depends on the river outflow. Rivers with large flows generally lead to more stratified
estuaries (U.S. EPA, 1985b).
Tidal height data and current predictions, published annually by NOAA, may provide
sufficient information, or it may be necessary to install a new tide gage (stage monitor) to develop
data closer to the CSO-impacted area. Due to the variation of tides and winds, estuarine and coastal
currents often change rapidly. It is necessary, therefore, to measure tides and currents
simultaneously using continuously recording depth and velocity meters. Tidal currents can be
measured with meters similar to those used for measurement of river currents, but the direction of
the currents must also be recorded. Information on monitoring methods for such areas may also be
found in USGS (1982) and USDI (1984).
Lakes
The hydraulic characteristics of lakes depend on several factors, including the depth, length,
width, surface area, volume, basin material, surrounding ground cover, typical wind patterns, and
surface inflows (including CSOs) and outflows. Lakes tend to have relatively low flow-through
velocities and significant vertical temperature gradients, and thus are usually not well-mixed
(Thomann and Mueller, 1987). To determine how quickly pollutants are likely to be removed from
a lake, it is necessary to define the flushing rate. The flushing rate depends on water inputs (inflows
and precipitation) and outputs (outflows, evaporation, transpiration, and withdrawal), pollutants and
their characteristics, and the degree of mixing in the lake. Mixing in lakes is primarily due to wind,
temperature changes, and atmospheric pressure.
Analysis of pollutant fate and transport in lakes is often complex and generally requires the
use of detailed simulation models. (Some less-complex analysis can be done, however, when
simplifying assumptions, such as complete mixing in the lake, are made.) Pollutant fate and
transport analysis requires definition of parameters such as lake volume, surface area, mean depth,
and mean outflow and inflow rates. Analytical and modeling methods for lakes and the data
necessary to use the methods are discussed in greater detail in Section 8.3.2 and in Thomann and
Mueller (1987) and Viessman, et al. (1977).
Receiving water hydraulic data can be analyzed to characterize the relationship between
depth, velocity, and flow in the receiving water. This analysis may involve:
l Evaluating the data to define hydraulic characteristics, such as initial dilution, mixing,
travel time, and residence time.
Plotting programs such as spreadsheets and graphics programs are useful for presenting
hydraulic data. A data base and a plotting and statistical analysis package will typically be necessary
to analyze the data and generate such information as:
l Plots of depth, velocity, and flow vs. distance from the outfall
l Means, standard deviations, and other important statistical measures for depth, velocity,
and flow data.
2
Stage-discharge curves, also referred to as rating curves, are plots of water level (stage) vs. discharge.
Development of these curves is discussed in Bedient and Huber (1992). USGS (1982) and USDI (1984)) discuss
methods for developing hydraulic curves for various types of flow monitoring stations.
As presented later in Chapter 8, receiving water models need physical system and hydraulic
data as input. Processing of input data is specific to each model. In general, however, the physical
characteristics of the receiving water (slopes, locations, and temperatures) are used to develop the
model computational grid. The measured hydraulic data (depths, velocities, and flows) are
compared with model calculations in order to validate the model.
Collection and analysis of receiving water quality data is necessary when available data are
not sufficient to describe water quality impacts from CSOs. This section discusses some approaches
for conducting receiving water sampling and for analyzing the collected samples. (Chapters 3 and
4 discuss how to identify sampling locations, sampling parameters, and sampling frequency. Section
6.4 discusses biological and sediment sampling and analysis.)
Receiving water monitoring involves many techniques similar to CSS monitoring (see
Section 5.4.1) and many of the same decisions, such as whether to collect grab or composite samples
and whether to use manual or automated methods. Receiving water quality monitoring involves the
parameters discussed in Section 4.5.3 as well as field measurement of parameters such as
temperature and conductivity.
Sampling receiving waters, especially large water bodies, requires careful planning and a
sizable resource commitment. For example, a dye study of a large river requires careful planning
regarding travel time, placement of sampling crews, points of access, safety concerns, and use of
boats. Sampling during wet weather events is typically more complicated than sampling during dry
weather, since it often requires rapid mobilization of several sampling teams on short notice,
sampling throughout the night, and sampling in rainy conditions with higher-than-normal flows in
the receiving water body. Time of travel between the various sampling stations may necessitate the
use of additional crews when sample collection must occur at predetermined times.
Wet weather sampling requires specific and accurate weather information. Local offices of
the American Meteorological Society can provide a list of Certified Consulting Meteorologists who
can provide forecasting services specific to the needs of a sampling program. Many weather services
also have Internet sites that provide real-time radar updates across the U.S. Radar coverage can also
be arranged in some areas for real-time observation of rainfall conditions. These efforts represent
an additional cost to the program, but they can be invaluable for planning wet weather surveys and
can result in significant savings in costs associated with false starts and unnecessary laboratory
charges. Section 4.6 discusses a strategy for determining whether to initiate monitoring for a
particular wet weather event.
The rainfall, darkness, and cold temperatures that often accompany wet weather field
investigations can make even small tasks difficult and sometimes unsafe. Contingency planning and
extensive preparation can, however, minimize mishaps and help ensure safety. Prior to field
sampling, the permittee should ensure that:
l Sampling personnel are well trained and familiar with their responsibilities, as defined
in the sampling plan
l A health and safety plan identifies the necessary emergency procedures, safety
equipment, and nearby rescue organizations and emergency medical services
l Sample containers are assembled and bottle labels are filled out to the extent possible
As discussed in Section 5.4.1, sample collection, preparation and handling, preservation, and
storage should minimize changes in the condition of sample constituents. The standard procedures
for collecting, preserving, and storing receiving water samples are the same as those for combined
sewage samples and are described in 40 CFR Part 136. Procedures for cleaning sample bottles,
preserving water quality samples, and analyzing for appropriate chemicals are detailed in various
methods manuals, including APHA (1992) and U.S. EPA (1979). NOAA’s Status and Trends
Program also provides information on standard protocols for sampling and analysis. Collection and
analytical methods depend on the constituents in the receiving water (e.g., salinity, suspended
sediments, ionic strength) as well as the required precision and accuracy. Samples should be labeled
with unique identifying information and should have chain-of-custody forms documenting the
changes of possession between time of collection and time of analysis (discussed in Section 5.4.1).
The use of sample bar code labels and recorders can be particularly helpful during wet weather
sampling when paper records are often infeasible.
As was the case for hydraulic data, water quality data for receiving waters are analyzed by
plotting and reviewing the raw data to define water quality characteristics and by processing the data
for input to water quality models. Data can be analyzed and displayed using spreadsheets, databases,
graphics software, and statistical packages, such as Statistical Analysis Software (SAS) and
3
Statistical Package for Social Sciences (SPSS).
l Comparing receiving water quality with applicable water quality criteria to determine
whether criteria are being exceeded
l Comparing sampling results from before, during, and after a wet weather event to assess
4
whether water quality problems are attributable to CSOs and other wet weather events
3
Use of these statistical packages generally requires solid statistical capabilities, so they should be used
cautiously by someone who is not experienced in statistical data evaluations and survey design. For information on
SAS, contact: SAS Institute, Inc., SAS Campus Drive, Cary, NC 27513 or (919) 677-8000. For information on
SPSS, contact: SPSS Incorporated, 444 N. Michigan Avenue, Chicago, IL 60611 or (800) 543-2185.
4
An alternative approach is to stratify the analysis by those samples that time travel analyses (e.g.,
Lagrangian analysis) indicate are impacted by CSO discharges. Many instream samples collected during a wet
weather event represent times either before or after the CSO “slug” has passed.
l Comparing data from downstream of CSOs to data collected upstream of CSOs (or to a
reference point) to distinguish CSO impacts.
Since CSO controls must ultimately provide for attainment of WQS, receiving water analyses
should be tailored to the applicable WQS. If the WQS for a pollutant contain numeric criteria
specifying frequency, magnitude, and duration, receiving water analyses should use the same
frequency, magnitude, and duration (see Sections 4.5, 9.1, and 9.2 for additional discussion.)
Water quality data are also used to calibrate receiving water models (see Chapter 8). This
is generally facilitated by plotting the data vs. time and/or distance to compare with model
simulations. Special studies may be required to determine rate constants, such as decay rates,
bacteria die-off rates, or suspended solids settling rates, if these values are used in the model.
It is often difficult and expensive to identify CSO impacts during wet weather using only
hydraulic and water quality sampling. Sediment and biological monitoring may be cost-effective
supplements or, in limited cases, alternatives to water quality sampling. Since sediment and
biological monitoring do not address public health risks, they can only be used as alternatives when
bacterial contamination is not a CSO concern. The following sections discuss sediment and
biological sampling techniques and data analysis.
Sediments are sinks for a wide variety of materials. Nutrients, metals, and organic
compounds bind to suspended solids and settle to the bottom of a water body when flow velocity is
insufficient to keep them in suspension. Once re-suspended through flood scouring, bioturbation,
desorption, or biological uptake, free contaminants can dissolve in the water column, enter sediment-
dwelling organisms, or accumulate or concentrate in fish and other aquatic organisms and
subsequently be ingested by humans and other terrestrial animals.
Typically, CSOs contain suspended material that can settle out in slower-moving sections
of receiving waters. Sediments can release accumulated contaminants for years after overflows have
been eliminated.
l The sampling device is lowered through the water column by a hand line or a winch
l The device is then activated either by the attached line or by a weighted messenger sent
down the line
l The scoops or jaws of the device close either by weight or spring action
5
Ideally, dredging should disturb the bottom as little as possible and collect all fine particles.
In cases where sediments are physically amendable to coring, core samples can be collected
to determine how pollutant types, concentrations, and accumulation rates have varied over time.
To avoid sample contamination, sediments should be removed from the dredge or core
sampler by scraping back layers in contact with the device and extracting sediments from the central
mass of the sample. In many cases the upper-most layer of sediment will be the most contaminated
and, therefore, of most interest. Sediment samples for toxicological and chemical examination
should be collected following Method E 1391 detailed in Standard Guide for Conducting Sediment
Toxicity Tests with Freshwater Invertebrates (ASTM, 1991).
5
Commonly used sediment samplers include the Ponar, Eckman, Peterson, Orange-peel, and Van Veen
dredges. Macroinvertebrate Field and Laboratory Methods for Evaluating the Biological Integrity of Surface
Waters (Klemm, 1990) has detailed descriptions of such devices.
Sediment data are typically analyzed by developing grain size distributions and plotting
concentrations of chemicals vs. distance. If the area of interest is two-dimensional horizontally,
isopleths can be plotted showing contours of constant concentration from the CSO outfall. If vertical
variations from core samples are available, concentration contours can also be plotted vs. depth.
Sediment chemistry data may be statistically analyzed to compare areas that are affected by CSOs,
non-CSO sources, and unaffected (background) areas. These analyses can give a longer-term view
of CSO impacts than water quality monitoring.
more useful for determining the overall impacts from all pollution sources on the biological health
of the receiving water.
Fish. Although other aquatic organisms may be more sensitive to pollutants, fish generate
the greatest public concern. Observable adverse effects from pollutants include declines of
populations and tumor growth on individuals. Fish monitoring programs can identify the relative
and absolute numbers of individuals of each species; the size distributions within species; growth
rates; reproduction or recruitment success; the incidence of disease, parasitism, and tumors; changes
in behavior; and the bioaccumulation of toxic constituents.
Common fish sampling methods include angling, seines, gill and trap nets, and
electrofishing. The references in Exhibit 6-1 provide guidance on methods used for collection,
measurement, preservation, and analysis of fish samples.’
6
Two reference works published by the American Fisheries Society are especially informative. Fisheries
Techniques (Nielsen and Johnson, 1983) focuses mainly on field work considerations, discussing most of the
sampling techniques currently practiced. The companion volume, Methods for Fish Biology (Schreck and Moyle,
1990) focuses primarily on methods used to analyze and assess collected fish samples. It includes material on fish
growth, stress and acclimation, reproduction, behavior, population ecology, and community ecology.
VanLandingham (1982) provide useful guides to the environmental requirements and pollution
tolerances of diatoms and blue-green algae, respectively.
Zooplankton. Zooplankton are free-floating aquatic protozoa and small animals. Many
species are sensitive indicators of pollution. Particularly in lakes and reservoirs, zooplankton can
provide information on the presence of specific toxics. Zooplankton are often collected by towing
a plankton net through a measured or estimated volume of water. To calculate population density
it is necessary to determine the volume of the sampling area, using a flow meter set in the mouth of
the net or calculations based on the area of the net opening and the distance towed. Laboratory
analyses can provide information similar to that for phytoplankton.
Monitoring teams generally use dredges, artificial substrates, and kicknets to sample benthic
macroinvertebrates, depending on the bottom substrate and water depth. Samples are either
preserved in their entirety in polyethylene bags or other suitable containers or are washed through
a fine sieve and then preserved in a suitable container (Klemm, 1990). The sample can be analyzed
for taxa present, the total density of each taxon, relative abundance by numbers or biomass of these
taxa, changes in major and indicator species populations, and the total biomass of benthic
7
macroinvertebrates present.
7
Three manuals (U.S. EPA, 1983b, 1984a, 1984b) discuss the interpretation of biological monitoring data
for larger bottom-living invertebrates. The Rapid Bioassessment Protocols (Plafkin et al., 1989) manual discusses
the use of fish and macroinvertebrates as a screening method in assessing environmental integrity.
Macroinvertebrate Field and Laboratory Methods for Evaluating the Biological Integrity of Surface Waters
(Klemm, 1990) discusses analysis of qualitative and quantitative data, community metrics and pollution indicators,
pollution tolerance of selected macroinvertebrates, and Hilsenhoff's family-level pollution tolerance values for
aquatic arthropods.
Community structure can be described in terms of species diversity, richness, and evenness.
Diversity is affected by colonization rates, the presence of suitable habitats, extinction rates,
competition, predation, physical disturbance, pollution, and other factors (Crowder, 1990).
A qualitative data assessment can help determine which factors have caused measured
variation in species diversity. In such an assessment, the collected species and their relative
population sizes are compared with their known sensitivities to contaminants present. The tendency
of species to be abundant, present, or absent relative to their tolerances or sensitivities to sediments,
temperature regimes, or various chemical pollutants can indicate the most likely cause of variation
in species diversity at the sampled sites.
Two cautions should be noted regarding qualitative analysis. First, different strains of the
same species can sometimes have differing sensitivities to a stressor, particularly where species have
undergone extensive hatchery breeding programs. Second, because listed characteristics of
organisms can vary from region to region, it is important when using lists of indicator species to note
whether the data were collected in the same region as the CSO study. Investigators should generally
limit the use of diversity indices as general indicators of environmental effects to comparisons within
the study where sampling and sample analysis methods are consistent. Before conducting a
biological assessment, investigators should contact local authorities to determine whether biological
reference data can be obtained to use in the CSO study. Data should be from biological reference
sites that have similar physical characteristics (e.g., comparable habitat).
Typically, RBPs provide integrated evaluations that compare habitat and biological measures
for studied systems to empirically-defined reference conditions (Plafkin et al., 1989). Reference
conditions are defined based on data from systematic monitoring of either a site-specific control
station or several comparable sites in the same region. A site-specific control is generally considered
to be representative of the “best attainable” conditions for a particular waterbody. When using data
from several regional sites, the sites are selected to represent the natural range of variation in “least
disturbed” water chemistry, physical habitat, and biological conditions. A percent similarity is
computed for each biological, chemical, or physical parameter measured at the study sites relative
to the conditions found at the reference site(s). These percentages may be computed based on the
total number of taxa found, dissolved oxygen saturation, or the embeddedness of bottom material.
Generally, where the computed percent similarity is greater than 75-80 percent of the
corresponding reference condition (depending on the parameter compared), the results can indicate
that conditions at the study sites are sufficiently similar to those occurring at the reference site(s).
For such cases it is reasonable to conclude that the study sites’ conditions are “non-impaired.” In
contrast, where the computed percent similarity of conditions at the study sites is less than 50 percent
of the reference conditions (depending on the parameter compared), it is reasonable to conclude that
conditions at those study sites are “severely impaired,” relative to the reference site(s). For those
sites with a percent similarity falling between these ranges, the results can indicate that conditions
at the study sites are “moderately impaired” (Plafkin et al., 1989). An application of the use of RBPs
in two case studies is presented in Combined Sewer Overflows and the Multimetric Evaluation of
Their Biological Effects: Case Studies in Ohio and New York (U.S. EPA, 1996).
CSS MODELING
This chapter discusses the use of modeling to characterize the combined sewer system (CSS)
and evaluate CSO control alternatives. It discusses different approaches to identifying the
appropriate level of modeling, based on site-specific considerations, and describes the various types
of available models. Because of the site-specific nature of CSSs, the varying information needs of
municipalities, and the numerous available models, it does not recommend a specific model or
modeling approach.
The CSO Control Policy refers to modeling as a tool for characterizing a CSS and the
impacts of CSOs on receiving waters. Although not every CSS needs to be analyzed using complex
computer models, EPA anticipates that most permittees will need to perform some degree of
modeling to support CSO control decisions.
weather events including the number, location and frequency of CSOs, volume,
concentration and mass of pollutants discharged and the impacts of the CSOs on the
receiving waters and their designated uses. (Section II.C.1)
The CSO Policy supports continuous simulation modeling (use of long-term rainfall records
rather than records for individual storms) for several reasons. Long-term continuous rainfall records
enable simulations to be based on a sequence of storms so that the additive effect of storms occurring
close together can be examined. They also enable storms with a range of characteristics to be
included. When a municipality uses the presumption approach, long-term simulations are
appropriate because the performance criteria are based on long-term averages, which are not readily
determined from design storm simulations. Continuous simulations do not require highly complex
models. Models that simulate runoff without complex simulation of sewer hydraulics (e.g.,
STORM, SWMM RUNOFF) may be appropriate where the basic hydraulics of the system are
simple or have been analyzed using a more complex model. In the second case, the results from the
more complex model can be used to enable proper characterization of system hydraulics in the
simple model.
Running a model in both continuous mode and single event mode can be useful for some
systems. When only long-term hourly rainfall data are available, it may be desirable to calibrate the
model using more refined single event rainfall data before running the model in continuous mode.
For instance, if a CSS is extremely responsive to brief periods of high-intensity rainfall, this may not
be adequately depicted using hourly rainfall data.
The CSO Control Policy also states that after instituting the nine minimum controls (NMC),
the permittee should assess their effectiveness and should
submit any information or data on the degree to which the nine minimum controls achieve
compliance with water quality standards (WQS). These data and information should include
results made available through monitoring and modeling activities done in conjunction with
the development of the long-term CSO control plan described in this Policy. (Section II.B)
The purpose of the system characterization, monitoring and modeling program initially is
to assist the permittee in developing appropriate measures to implement the nine minimum
controls and, if necessary, to support development of the long-term CSO control plan. The
monitoring and modeling data also will be used to evaluate the expected effectiveness of both
the nine minimum controls, and, if necessary, the long-term CSO controls, to meet WQS.
(Section II.C.1)
The long-term control plan (LTCP) should be based on more detailed knowledge of the CSS
and its receiving waters than is necessary to implement the NMC. The LTCP should consider a
reasonable range of alternatives, including various levels of controls. Hydraulic modeling may be
necessary to predict how a CSS will respond to various control scenarios. A computerized model
may be necessary for a complex CSS, especially one with looped networks or sections that
surcharge. In simpler systems, however, basic equations (e.g., Hazen-Williams or Manning equation
- see Section 5.3.1) and spreadsheet programs can be used to compute hydraulic profiles and predict
the hydraulic effects of different control measures. (Verification using monitoring data becomes
more important in these latter situations.)
Finally, modeling can support either the presumption or demonstration approaches of the
CSO Control Policy. The demonstration approach requires demonstration that a proposed LTCP
is adequate to meet CWA requirements. Meeting this requirement can necessitate detailed CSS
modeling as an input to receiving water impact analyses. On the other hand, the presumption
approach involves performance-based limits on the number or volumes of CSOs. This approach
may require less modeling of receiving water impacts, but is acceptable only if “thepermitting
authority determines that such presumption is reasonable in light of the data and analysis conducted
in the characterization, monitoring, and modeling of the system and the consideration of sensitive
areas . . . .” (Section II.C.4.a) Therefore, the presumption approach does not eliminate the need to
consider receiving water impacts.
This section discusses how to select a CSS model. Generally, the permittee should use the
simplest model that meets the objectives of the modeling effort. Although complex models usually
provide greater precision than simpler models, they also require greater expense and effort. This
section does not describe all of the available CSS-related models, since other documents provide this
information (see Shoemaker et al., 1992; Donigian and Huber, 1991; WPCF, 1989).
l Hydrology is the key factor in determining runoff in CSS drainage basins. Hydrologic
modeling is generally done using runoff models to estimate flows influent to the sewer
system. These models provide input data for hydraulic modeling of the CSS.
l CSS hydraulic modeling predicts the pipe flow characteristics in the CSS. These
characteristics include the different flow rate components (sanitary, infiltration, inflow,
and runoff), the flow velocity and depth in the interceptors, and the CSO flow rate and
duration.
l CSS water quality modeling consists of predicting the pollutant characteristics of the
combined sewage in the system, particularly at CSO outfalls and at the treatment plant.
CSS water quality is measured in terms of bacterial counts and concentrations of
important constituents such as BOD, suspended solids, nutrients, and toxic contaminants.
Since hydraulic models are usually used together with a runoff model or have a built-in
runoff component, runoff models are discussed as part of hydraulic modeling in the following
sections.
Some models include both hydraulic and water quality components, while others are limited
to one or the other. Although CSO projects typically involve hydraulic modeling, water quality
modeling in the CSS is less common, and a community may decide to rely on CSS water quality
monitoring data instead.
Several factors will dictate whether CSS water quality modeling is appropriate. WPCF
(1989) concludes that “simulation of quality parameters should only be performed when necessary
and only when requisite calibration and verification data are available[...] Another option is to couple
modeled hydrologic and hydraulic processes with measured quality data to simulate time series of
loads and overflows.” Modeling might not be justified in cases where measured CSS water quality
variations are difficult to relate to parameters such as land use, rainfall intensity, and pollutant
accumulation rates. For these cases, using statistics (such as mean and standard deviation) of CSS
water quality parameters measured in the sewer system can be a valid approach. One limitation of
this approach, however, is that it cannot account for the implementation of best management
practices (BMPs) such as street sweeping or the use of detention basins.
Exhibit 7-1 shows how model selection can be affected by the status of NMC implementation
and LTCP development, and by whether the LTCP will be based on the presumption or
demonstration approach. To avoid duplication of effort, the permittee should always consider
modeling needs that will arise during later stages of LTCP development or implementation.
approach, this information will help determine the amount and timing of pollutant loadings to the
receiving water.
Demonstration Approach. Under the demonstration approach, the permittee needs to show
that the planned controls will provide for attainment of WQS unless WQS cannot be attained as a
result of natural background conditions or pollution sources other than CSOs.
Therefore, CSS modeling under the demonstration approach should describe pollutant
loadings to the receiving water body. Since water quality modeling in the CSS is directly linked to
water quality modeling in the receiving water, the CSS model must generate sufficient data to drive
the receiving water model. Further, the resolution needed for the CSS pollutant transport estimates
will depend on the time resolution called for in the receiving water model, which is in turn driven
by WQS. For pollutants with long response times in the receiving water (such as BOD and
nutrients), the appropriate level of loading information is usually the total load introduced by the
CSO event. For pollutants with shorter response times (such as bacteria and acutely toxic
contaminants), it may be necessary to consider the timing of the pollutant load during the course of
the CSO event.
Hydraulic models used for CSS simulations can be divided into three main categories:
1
l Runoff models based on Soil Conservation Service (SCS) runoff curve numbers, runoff
coefficients, or other similar methods for the generation of flow. These models can
estimate runoff flows influent to the sewer system and, to a lesser degree, flows at
different points in the system. Runoff models do not simulate flow in the CSS, however,
and therefore do not predict such parameters as the flow depth, which frequently control
the occurrence of CSOs. (The RUNOFF block of EPA’s Storm Water Management
2
Model (SWMM) is an example. )
1
SCS runoff curves were developed based on field studies measuring runoff amounts from different soil cover
combinations. The appropriate runoff curve is determined from antecedent moisture condition and the type of soil.
(Viessman et al., 1977)
2
The SWMM RUNOFF model also has limited capabilities for flow routing in the CSS.
3
Flow, which is caused by the motion of waves, can be described by the hydraulic routing technique. This technique
is based on the simultaneous solution of the fully hydrodynamic equations (the continuity equation and the momentum
equation for varying flow). Under certain conditions, these hydrodynamic equations can be simplified to a one-
dimensional continuity equation and a uniform flow equation (in place of the full momentum equation). This is referred
to as the kinematic wave approximation (discharge is simply a function of depth). (Bedient and Huber, 1992)
These models require the user to input hydrographs from runoff model results. (The
TRANSPORT block of SWMM is an example.)
l Complex, dynamic models based on the full hydrodynamic equations. They can
simulate surcharging, backwater effects, or looped systems, and represent all pertinent
processes. These models require the user to input hydrographs from runoff model
results. (The EXTRAN block of SWMM is an example.)
Exhibit 7-2 compares the flow routing capabilities of the three SWMM blocks. Section 7.3 discusses
available hydraulic models.
The simpler models were developed to support rapid evaluations of CSSs. They require little
input data, are relatively easy to use, and require less computer time than complex models. These
features, however, are becoming less significant because complex models with user-friendly pre- and
post-processors are now widely available. Advances in computer technology render run-time a
secondary issue for all but the largest of applications.
2. Ability to accurately represent runoff in the CSS drainage basin. The runoff
component of the hydraulic model (or the runoff model, if a separate hydrologic
model is used) should adequately estimate runoff flows influent to the sewer system.
It should adequately characterize rainfall characteristics as well as hydrologic factors
such as watershed size, slope, soil types, and imperviousness.
1
After Huber and Dickinson, 1988.
2
Backwater may be simulated as a horizontal water surface behind a storage element.
3
The RUNOFF block sub-model is primarily intended to calculate surface runoff, but includes the capability to simulate
simple channel conveyance of flows. The TRANSPORT and EXTRAN blocks are sewer conveyance models with no
runoff components and thus require user input of hydrographs.
most of these models use a complex finite-difference technique to solve for the
governing equations. Sound simulation of hydraulic behavior requires that the
modeler achieve numeric stability of the solution technique through the selection of
appropriate time and space intervals. In some cases, however, estimates of overflow
at unmonitored locations can be made based on monitoring in areas with similar
geographic features (like slope, degree of imperviousness, or soil conditions), based
4
on V/R ratios and drainage basin characteristics (see Section 5.3.3).
5. Need to assess water quality in CSS. If CSS water quality simulations are needed,
the permittee should consider the model’s capability to simulate water quality. To
simulate CSS water quality, it is often better to use actual pollutant concentrations
from monitoring results together with modeled CSS flows.
6. Need to assess water quality in receiving waters. The pollutants of concern and
the nature of the receiving water affect the resolution of the CSO data needed for the
water quality analyses. For example, bacteria analysis typically requires hourly
rather than daily loading data, and the hydraulic model must be capable of providing
this resolution.
4
V/R is the ratio of the overflow volume to the rainfall depth.
This in turn will constrain the time resolution for CSS hydraulic modeling. The
permittee should consider the level of time resolution derived when selecting a
model.
9. Ease of use and cost. As mentioned above, simple models tend to be easier to use
than complete dynamic models. Although user-friendly dynamic models now exist,
they are generally commercial models that cost more than public domain models and
can be used incorrectly by inexperienced users. Another option is to use commercial
pre- and post-processors (or shells) designed to facilitate the use of public domain
models such as SWMM. They can provide graphically-oriented, menu-driven data
entry and extensive results plotting capabilities at a cost lower than that of complete
dynamic models.
Another issue related to ease of use and accuracy is robustness, which is a model’s
lack of propensity to become unstable. Instabilities are uncontrolled oscillations of
the model’s results due to the approximations made in the numerical solution of the
basic differential equations. Instabilities tend to occur primarily in fully dynamic
models, and are caused by many factors, including incomplete sewer information and
short conduits. Resolving model instabilities can be time-consuming and requires
extensive experience with the model.
CSS water quality models can be divided into the following categories:
l
Land Use Loading Models - These models provide pollutant loadings as a function of
the distribution of land uses in the watershed. Generally, these models attribute to each
land use a concentration for each water quality parameter, and calculate overall runoff
quality as a weighted sum of these concentrations. Pollutant concentrations for the
different land uses can be derived from localized data bases or the Nationwide Urban
Runoff Program (NURP), a five-year study initiated in 1978 (U.S. EPA, 1983a). Local
data are usually preferable to NURP data since local data are generally more recent and
site-specific.
l Build-Up/Washoff Models - These models simulate the basic processes that control
runoff quality, accounting for such factors as time periods between events, rainfall
intensity, and BMPs. They require calibration and are not regularly used due to the
expense and difficulty of defining site-specific rates.
Many models do not address the potentially important role of chemical reactions and
transformations within the CSS. Calibration may be difficult because pollutant loading into the CSS
is often uncertain.
The permittee should consider the following criteria when selecting a CSS water quality
model:
1. Needs of the receiving water quality simulation. The time scale of the pollutant
concentration simulation in the CSS, and the degree of sophistication of the model,
depends partly on the needs of the receiving water quality simulation (if used) and,
ultimately, on the level of detail required to demonstrate attainment of WQS. If it is only
necessary to estimate the average annual loading to the receiving water, then detailed
hourly or sub-hourly simulation of combined sewage quality generally will not be
necessary. As noted above, in many cases it is appropriate to combine sophisticated
hydraulic modeling with approximate CSS water quality modeling.
2. Ability to assess control and BMP alternatives. When the control alternatives under
assessment include specific BMPs or control technologies, the CSS water quality model
should be sophisticated enough to estimate the effects of these alternatives.
4. Capability for pollutant routing. Another concern is the model’s capability for
pollutant routing-i.e., its capacity to account for variability in pollutant concentrations
during storm events. Most models translate pollutant concentrations from sources and
CSO quantity to pollutant loading without taking separate account of the timing of
pollutant delivery due to transport through the CSS. Some basins deliver the highest
concentrations of pollutants in the rising limb of the storm flow (the “first flush” effect).
If the CSO loading for such systems is modeled using overflow quantity and average
concentrations, inaccuracies may result, particularly if the “first flush” is effectively
captured by the POTW or storage.
5. Expense and ease of use. Sophisticated water quality models can be expensive to
calibrate and generally are more difficult to use. If a simpler model is applicable to the
situation and can be properly calibrated, it may be sufficient and can be more accurate.
Exhibit 7-3 summarizes several runoff and hydraulic models and Exhibit 7-4 summarizes
several water quality models. These models have been developed by EPA and the Army Corps of
Engineers and are available in the public domain. Some of the models in Exhibit 7-3 are runoff
models (such as STORM); others have a runoff component but also simulate flow in the CSS (such
as SWMM and Auto-Q-ILLUDAS).
5
The commercial packages have not been reviewed by EPA and they are subject to continued evolution and change,
like all commercial software. This listing is provided to assist potential users; it is not meant to endorse any particular
model or imply that models not listed are not acceptable. A recent listing of some available models is found in Mao
(1992). Recent developments in sewer and runoff models include linking models to geographic information systems
(GIS), computer-aided design (CAD) systems, and receiving water models such as WASP.
Some of the public domain models listed above are available from EPA’s Center for
Exposure Assessment Modeling (CEAM). CEAM can be contacted at:
CEAM
National Exposure Research Laboratory-Ecosystems Research Division
Office of Research and Development
USEPA
960 College Station Road
Athens, GA 30605-2700
Voice: (706) 355-8400
Fax: (706) 355-8302
e-mail: ceam@epamail.epa.gov
CEAM also has an Internet site at http://www.epa.gov/CEAM/
Some of the public domain models listed above are available from EPA’s Center for
Exposure Assessment Modeling (CEAM). CEAM can be contacted at:
CEAM
National Exposure Research Laboratory-Ecosystems Research Division
U.S. EPA Office of Research and Development
960 College Station Road
Athens, GA 30605-2700
Voice: (706) 355-8400 Fax: (706) 355-8302
e-mail: ceam@epamail.epa.gov
CEAM also has an Internet site at http://www.epa.gov/CEAM/
January 1999
Chapter 7 CSS Modeling
Exhibit 7-5. Selected Commercial CSS Models
Type of Hydraulic Water Quality
Package Name Simulation Capability Contact
Hydra/Hydra Graphics Dynamic No PIZER Incorporated
4422 Meridian Avenue N
Seattle, Washington 98103
(800) 222-5332
www.pizer.com
6
example, SWMMDuet allows the integration of SWMM and Arc/INFO for database management
and GIS analysis.
These exhibits summarize some important technical criteria, and can be used as a preliminary
guide. However, to evaluate the use of a specific model in a particular situation the permittee should
refer to the more detailed reviews and major references listed in Exhibits 7-3 and 7-4. Both
Shoemaker et al. (1992) and Donigian and Huber (1991) provide preliminary evaluations of the
functional criteria, including cost and data requirements. The Water Resources Handbook (Mays,
1996) discusses both hydraulic and water quality models and compares their attributes.
6
SWMMDuet is a SWMM/GIS Interface. Further information can be obtained from the Delaware Department of
Natural Resources at (302) 739-3451.
In developing the model, the modeler establishes initial conditions for various model
components (such as the level of discretization) and input data parameters (such as percent
imperviousness of subcatchments). These elements are then adjusted through model calibration,
which is discussed in the next section.
Until recently the modeler had to compromise between the level of detail in a model
(temporal and spatial precision), the mode in which it was run (complex vs. simple), and the time
period for the simulation (event vs. continuous). As computer technology continues to improve,
limitations in computing power are becoming less of a factor in determining the appropriate level
of modeling complexity. However, for increased model complexity to lead to greater accuracy,
complex models should be used by knowledgeable, qualified modelers who have sufficient
supporting data. In some cases, where detail is not required, a simplified model may save time spent
filling the data requirements of the model, preparing tiles, and doing the model runs. Shoemaker
et al. (1992, Tables 7 to 9) provides a tabular summary of the main input and output data for each
of the models presented in Exhibits 7-3 and 7-4.
The level of discretization (i.e., coarse vs. fine scale) determines how precisely the geometry
of the CSS and the land characteristics of the drainage basin are described in the model. At a very
coarse level of discretization, the CSS is a black box with lumped parameters and the model (e.g.,
STORM) primarily simulates CSOs. A more complex approach might be to simulate the larger
pipes of the CSS, but to lump the characteristics of the smaller portions of the CSS. Another
intermediate level of complexity is to simulate the interceptor when it is the limiting component in
the CSS for controlling overflows. Much can be learned about system behavior by simulating
interceptor hydraulics in response to surface runoff. More complex simulations would include
increasing levels of detail about the system.
For systems that are controlled hydraulically at their downstream ends, it may only be necessary to
model the larger downstream portion of the CSS. If flows are limited due to surcharging in upstream
areas, however, a simulation neglecting the upstream portion of the CSS would over-estimate flows
in the system. In some cases it is difficult to determine ahead of time what the appropriate level of
detail is. In these cases, the modeler can take an incremental approach, determining the value of
additional complexity or data added at each step. Exhibit 7-6, for example, compares a simulation
based on five subcatchments (coarse discretization) and a simulation based on twelve subcatchments
(finer discretization) with observed values. Only marginal improvement is observable when
subcatchments are increased from five to twelve. The modeler should probably conclude that even
finer discretization (say, 15 subcatchments) would provide little additional value.
A model general enough to tit a variety of situations typically needs to be adjusted to the
characteristics of a particular site and situation. Model calibration and validation are used to “fine-
tune” a model to better match the observed conditions and demonstrate the credibility of the
simulation results. An uncalibrated model may be acceptable for screening purposes, but without
supporting evidence the uncalibrated result may not be accurate. To use model simulation results
for evaluating control alternatives, the model must be reliable.
Calibration is the process of running a model using a set of input data and then comparing
the results to actual measurements of the system. If the model results do not reasonably approximate
actual measurements, the modeler reviews the components of the model to determine if adjustments
7
should be made so that the model better reflects the system it represents. For example, a CSS
hydraulic model used to simulate overflows is calibrated by running the model using measured
rainfall data to simulate the volume, timing, and depth of CSOs. The model results are then
compared to actual measurements of the overflows. The modeler then adjusts parameters such as
the Manning roughness coefficient or the percent imperviousness of subcatchments within
scientifically credible ranges and runs the model a second time, again comparing the results to
observations. Initial calibration runs often point to features of the system, such as a connection or
bypass, which may not have been evident based on the available maps. The modeler repeats this
procedure until satisfied that the model produces reasonable simulations of the overflows. Models
are usually calibrated for more than one storm, to ensure appropriate performance for a range of
conditions. Exhibit 5-9 shows some example model calibration plots of flow and depth during storm
events. For calibration, the most important comparisons are total volumes, peak flows, and shapes
of the hydrographs.
Validation is the process of testing the calibrated model using one or more independent data
sets. In the case of the hydraulic simulation, the model is run without any further adjustment using
independent set(s) of rainfall data. Then the results are compared to the field measurements
collected concurrently with these rainfall data. If the results are suitably close, the model is
considered to be validated. The modeler can then use the model with other sets of rainfall data or
at other outfalls. If validation fails, the modeler must recalibrate the model and validate it again
using a third independent data set. If the model fails a validation test, the next test must use a new
data set. (Re-using a data set from a previous validation test does not constitute a fair test, because
the modeler has already adjusted model parameters to better fit the model to the data.) Validation
is important because it assesses whether the model retains its generality; that is, a model that has
been adjusted extensively to match a particular storm might lose its ability to predict the effects of
other storms.
7
Model calibration is not simply “curve fitting” to meet the data. Model adjustments should make the modeled
elements of the system better reflect the actual system.
The availability of adequate calibration data places constraints on which models are
appropriate. When identifying the time period for conducting CSS flow monitoring, the permittee
should consider the effect of using larger data sets. The Combined Sewer Overflow Control Manual
(U.S. EPA, 1993) states that “an adequate number of storm events (usually 5 to 10) should be
monitored and used in the calibration.” The monitoring period should indeed cover at least that
many storms, but calibration and validation are frequently done with 2 to 3 storms each.
Most models are more accurate when applied in a relative rather than an absolute manner.
Model output data concerning the relative contribution... to overall pollutant loads is more
reliable than an absolute prediction of the impacts of one control alternative viewed alone.
When examining model output. . . it is important to note three factors that may influence the
model output and produce unreasonable data. First, suspect data may result from
calibration or verification data that are insufficient or inappropriately applied. Second, any
given model, including detailed models, may not represent enough detail to adequately
describe existing conditions and generate reliable output. Finally, modelers should
remember that all models have limitations and the selected model may not be capable of
simulating desired conditions. Model results must therefore be interpreted within the
limitations of their testing and their range of application. Inadequate model calibration and
verification can result in spurious model results, particularly when used for absolute
predictions. Data limitations may require that model results be used only for relative
comparisons.
Common practice employs both judgment and graphical analysis to assess a model’s
adequacy. However, statistical evaluation can provide a more rigorous and less subjective approach
to validation (see Reckhow et al., 1990, for a discussion of statistical evaluation of water quality
models).
Nix ( 1990) suggests the following general sequence for calibrating a CSS model:
2. Classify model parameters to determine the degree to which they can be directly
measured, or, alternatively, are conceptual parameters not amenable to direct
measurement. For instance, a parameter such as area is usually easily defined, and thus
not varied in calibration, while parameters that are both important to model performance
and not amenable to direct measurement (e.g., percent imperviousness) will be the
primary adjustment factors for calibration.
3. Calibrate the model first for the representation (prediction) of overflow volume.
5. Finally, calibrate the pollutant parameters only after an acceptable flow simulation
has been obtained.
Once a model has been calibrated and validated, it can be run for long-term simulations
and/or for single events (usually a set of design storms).
l Long-term simulations can account for the sequencing of the rainfall in the record and
the effect of having storms immediately follow each other. They are therefore useful for
assessing the long-term performance of the system under the presumption approach.
Long-term simulations also assess receiving water quality accurately under the
demonstration approach. Water quality criteria need to be evaluated with the frequency
and duration of exceedance in order to be relevant. This is best done using long-term
continuous simulations or skillfully done probabilistic simulations. Although continuous
simulation models should be calibrated using continuous data where possible, they may
be calibrated with single events if antecedent conditions are taken into account. As the
l Single event simulations are useful for developing an understanding of the system
(including the causes of CSOs) and formulating control measures, and can be used for
calibrating models.
Although increased computer capabilities enable continuous simulations with greater levels
of detail, continuous simulation of very large systems can have some drawbacks:
l The model may generate so much data that analysis and interpretation are difficult
l Limitations in the accuracy of hydrologic input data (due to the inability to continuously
simulate spatially variable rainfall over a large catchment area) may lead to an inaccurate
time series of hydraulic conditions within the interceptor
l The more storms that are simulated, the greater the chance that instabilities will occur in
complex models. Correctly identifying and resolving these instabilities requires capable,
experienced modelers.
Model Output
The most basic type of model output is text files in which the model input is repeated and
the results are tabulated. These can include flow and depth versus time in selected conduits and
junctions, as well as other information, such as which conduits are surcharging. The model output
may include an overall system mass balance with such measures as the runoff volume entering the
system, the volume leaving the system at the downstream boundaries, the volume lost due to
flooding, and the change of volume in storage. The model output can also measure the mass balance
accuracy of the model run, which may indicate that problems, such as instabilities (see
Section 7.2.1) occurred.
Most models also produce plot tiles, which are easier to evaluate than text files. Output data
from plot files can be plotted using spreadsheet software or commercial post-processors, which are
available for several public domain models (particularly SWMM). Commercial models typically
include extensive post-processing capabilities, allowing the user to plot flow or depth versus time
at any point in the system or to plot hydraulic profiles versus time along any set of conduits.
Interpretation of Results
Simulation models predict CSO volumes, pollutant concentrations, and other variables at a
resolution that depends on the model structure, model implementation, and the resolution of the
input data. Because the ultimate purpose of modeling is generally to assess the CSO controls needed
to provide for the attainment of WQS, the model’s space and time resolution should match that of
the applicable WQS. For instance, a State WQS may include a criterion that a one-hour average
concentration not exceed a given concentration more than once every 3 years on average. Spatial
averaging may be represented by a concentration averaged over a receiving water mixing zone, or
implicitly by the specification of monitoring locations to establish whether the instream criteria can
be met. In any case, the permittee should note whether the model predictions use the same averaging
scales as the relevant water quality criteria. When used for continuous rather than event simulation,
as suggested by the CSO Control Policy, simulation models provide output that can be analyzed to
predict the occurrence and frequency of water quality criteria exceedances.
In interpreting model results, the permittee needs to be aware that modeling usually will not
provide exact predictions of system performance measures such as overflow volumes or exceedances
of water quality criteria. With sufficient effort, the permittee often can obtain a high degree of
accuracy in modeling the hydraulic response of a CSS, but results of modeling pollutant
buildup/washoff, transport in the CSS, and fate in receiving waters are considerably less accurate.
Achieving a high degree of accuracy may be more difficult in a continuous simulation because of
the difficulty of specifying continually changing boundary conditions for the model parameters.
l Model predictions are only as accurate as the user’s understanding and knowledge of the
system being modeled and the model being used
l Model predictions are no better than the quality of the calibration and validation exercise
and the quality of the data used in the exercise
l Model predictions are only estimates of the response of the system to rainfall events.
Evaluating these issues involves the closely related concepts of model accuracy and
reliability. Accuracy is a measure of the agreement between the model predictions and
observations. Reliability is a measure of confidence in model predictions for a specific set of
conditions and for a specified confidence level. For example, for a simple mean estimation, the
accuracy could be measured by the sample standard deviation, while the reliability of the prediction
(the sample mean in this case) could be evaluated at the 95 percent confidence level as plus or minus
approximately two standard deviations around the mean.
Modeling as part of LTCP development enables the permittee to demonstrate that a given
control option is “likely” to result in compliance with the requirements of the CWA and attainment
of applicable WQS. During LTCP development, the permittee will justify that a proposed level of
control will be adequate to provide for the attainment of WQS. Therefore, the permittee should be
prepared to estimate and document the accuracy and reliability of model predictions.
Evaluating model accuracy and reliability is particularly important for the analysis of
wet-weather episodic loading, such as CSOs. Such analysis invariably involves estimation of
duration (averaging period) and frequency of excursion above a water quality criterion, regardless
of whether the criterion is expressed as average monthly and maximum daily values, or as a
maximum concentration for a given design stream flow (e.g., 7Q10). Estimating duration and
frequency of excursion requires knowledge of model reliability, and the duration and frequency of
the storm events serving as a basis for the model.
l Sensitivity analysis is the simplest and most commonly used technique in water quality
modeling (U.S. EPA, 1995g). Sensitivity analysis assesses the impact of the uncertainty
of one or more input variables on the simulated output variables.
l
Monte Carlo simulation, a more complex technique, is a numerical procedure where
an input variable is defined to have a certain probability density function (pdf). Before
each model run, an input variable is randomly selected from each predefined pdf. By
combining the results of several model runs, a pdf can be developed for the output
variable which is useful in predicting overall model results. The number of model runs
is extremely large compared to the number of runs typically done for sensitivity or first-
order error analysis. Monte Carlo analysis can be used to define uncertainty (due to
uncertain model coefficients) and environmental variability (using historical records to
characterize the variability of inputs such as stream flow).
The main input variables for simulating the impact of CSO loadings are properties of the
mean rainfall event (storm event depth, duration, intensity, and interval between events), CSO
concentrations of specific pollutants, design flow of the receiving water body, and its background
8
concentrations. The output consists of an assessment of the water quality impact in terms of
duration and frequency of exceedances of water quality criteria. CSO pollutant concentrations are
the main “uncertain” (sensitive) input variables and can be varied over a range of reasonable values
to assess their impact on the resulting water quality. Uncertainty analysis can improve management
decisions and indicate the need for any additional data collection to refine the estimated loads. For
instance, if a small change in CSO pollutant concentrations results in an extremely large variation
in the prediction of water quality, it may be appropriate to allocate resources to more accurately
estimate the CSO pollutant concentrations used in the model.
This section applies the Storm Water Management Model (SWMM) to a single drainage area
from the example CSS drainage area presented in Chapters 4 and 5. While some of the details of
the application are particular to the SWMM model, most of the explanation applies to a range of
hydraulic models. The TRANSPORT block of the SWMM model was chosen for the flow routing
because the system hydraulics did not include extensive surcharging, and the system engineers felt
that a dynamic hydraulic model such as SWMM EXTRAN was not needed to accurately predict the
number and volume of CSOs.
The first step in model application is defining the limits of the combined sewer service area
and delineating subareas draining to each outfall (see Exhibit 7-7). This can be done using a sewer
system map, a topographic map, and aerial photographs as necessary. The modeler next must decide
what portions of the system to model based on their contributions to CSOs (as illustrated in
Example 4-1). The modeler then divides selected portions of the CSS and drainage area into
segments and translates drainage area and sewer data into model parameters. This process, referred
8
Continuous simulations do not require use of the “mean” rainfall event or “design” flow data.
Not to Scale
Chapter 7 CSS Modeling
to as discretization, begins with the identification of drainage boundaries, the location of major sewer
inlets using sewer maps, and the selection of channels and pipes to be represented in the model. The
drainage area is then further divided into subareas, each of which contributes to the nodes of the
simulated network.
The modeler must consider the tradeoff between a coarse model that simulates only the
largest structures in the CSS, and a fine-scale model that considers nearly every portion of the CSS.
A coarse model requires less detailed knowledge of the system, less model development time, and
less computer time. The coarse model, however, leaves out details of the system such as small pipes
and structures in the upstream end of the CSS. Flow in systems that are limited by upstream
structures and flow capacities will not be simulated accurately.
Where pipe capacities limit the amount of flow leaving a drainage area or delivered to the
wastewater treatment plant, the modeler should use the flow routing features of the model to
simulate channels and pipes in those areas of concern. The level of detail should be consistent with
the minimum desired level of flow routing resolution. For example, information cannot be obtained
about upstream storage unless the upstream conduits and their subcatchments are simulated. Further,
sufficient detail needs to be provided to allow control options within the system to be evaluated for
different areas.
In this example, the modeled network is carried to points where the sewers branch into pipes
smaller than 21 inches. The system is not directly modeled upstream of these points. Instead, runoff
from the upstream area is estimated and routed into the 21-inch pipes. Exhibit 7-8 shows the
modeled sewer lines and the subareas tributary to those lines for Service Area 1.
RUNOFF block. The RUNOFF block of SWMM generates surface runoff and pollutant
loads in response to precipitation input and modeled surface pollutant accumulations. The main data
inputs for the RUNOFF block are:
l
subcatchment width
l
subcatchment area
l
subcatchment imperviousness
l
subcatchment ground slope
l Manning’s roughness coefficient for impervious and pervious areas
l
impervious and pervious area depression storage
l
infiltration parameters.
Exhibit 7-9 shows the main RUNOFF block data inputs (by subcatchment area number) for the
example. The subcatchment area is measured directly from maps. Subcatchment width is generally
measured from the map, but is more subjective when the subcatchment is not roughly rectangular,
symmetrical and uniform. Slopes are taken from topographic maps, and determinations of
imperviousness, infiltration parameters, ground slope, Manning’s roughness coefficients, and
depression storage parameters are based on field observations and aerial photographs.
The RUNOFF block data file is set up to generate an interface file that transfers hydrographs
generated by the RUNOFF block to subsequent SWMM blocks for further processing. In this
example, the data generated in the RUNOFF block are processed by the TRANSPORT block.
TRANSPORT block. The TRANSPORT block is typically used to route flows and
pollutant loads through the sewer system. TRANSPORT also allows for the introduction of dry
weather sanitary and infiltration flow to the system. Exhibit 7-10 presents the main TRANSPORT
block inputs by element number. It lists the number and type of each element (including upstream
elements), the element length (for pipe elements), and inflow (for manholes).
Inflow (cfs)
ement Data [for manhole] Pipe Manning Pipe
Upstream Upstream Element or Length (ft) Dimension Pipe Slope Roughness
No. Element No. 2 Element No. 3 Type [for pipe element] (ft) (ft/10 ft) (n)
125 175 0 0 manhole 0.087 NA NA1 NA1
175 126 0 0 sewer pipe 1000 .45 0.5 0.014
126 176 177 0 manhole 0.188
177 150 0 0 sewer pipe 840 2.75 0.28 0.014
150 178 179 0 manhole 0
178 127 0 0 sewer pipe 390 1.75 0.39 0.014
127 0 0 0 manhole 0.097
179 128 0 0 sewer pipe 651 2.0 0.34 0.014
128 0 0 0 manhole 0.163
176 129 0 0 sewer pipe 733 4.5 0.07 0.014
129 180 0 0 manhole 0.076
180 130 0 0 sewer pipe 841 4.0 0.16 0.014
130 181 0 0 manhole 0.176
181 131 0 0 sewer pipe 620 4.0 0.09 0.014
131 182 0 0 manhole 0.136
182 132 0 0 sewer pipe 727 3.5 0.12 0.014
132 183 0 0 manhole 0.103
183 133 0 0 sewer pipe 771 3 0.16 0.014
133 184 0 0 manhole 0.221
184 134 0 0 sewer pipe 1110 2.75 0.13 0.014
134 185 0 0 manhole 0.258
185 135 0 0 sewer pipe 1007 1.75 0.4 0.014
135 0 0 0 manhole 0.131
1
Parameter is not applicable for manholes.
Chapter 7 CSS Modeling
The inflow parameter allows for introduction of dry-weather (sanitary) flow to the system.
Dry-weather flow is typically distributed proportional to area served. Here it is set to 0.0035 cfs per
acre. If the records are available, this parameter can be refined by multiplying the per-capita
wastewater flow (typically available from the wastewater treatment plant or latest facilities plan) by
the average population density calculated from census figures and sewer service area maps.
Exhibit 7-11 shows the output hydrograph for element (manhole) 125 from the
TRANSPORT block, with the measured flow for the event plotted for comparison. The peak flow,
shape of the hydrograph, and the total volume of overflow for thiscalibration run are very close to
the measured values.
The SWMM model is applied to monitored drainage areas within the CSS using available
monitoring data to calibrate the hydraulic portions of the program to monitored areas. For outfalls
that are not monitored, parameters are adjusted based on similar monitored areas and on flow depths
or flow determinations obtained from the initial system characterization (see Chapter 3). Once the
entire CSS drainage area is modeled and the SWMM model calibrated, the model then needs to be
validated. It can then be used to predict the performance of the system for single events (actual or
design) and/or for a continuous rainfall record. Recall that it is desirable to calibrate the model to
a continuous sequence of storms if is to be applied to a continuous rainfall record. Individual storms
related to monitored events can be run to calculate the total volume of overflow for the system. Peak
flow values from the SWMM hydrographs can be used for preliminary sizing of conveyance
facilities that may be needed to alleviate restrictions.
To predict the number of overflows per year, the calibrated model can be run in a continuous
mode and/or for design storm events. In the continuous mode the model can be run using the long-
term rainfall record (preferable where the data are available), or for a shorter period of time (e.g., for
a typical or extreme year from the example discussed throughout Chapter 5). While the event mode
is useful for some design tasks and for estimating hourly loading for a fine-scale receiving water
model, the continuous mode is preferable for evaluating the number of overflows under the
presumption approach. In this example, the model was run in continuous mode, using data from the
Predicted Measured
38-year rainfall record. The model predicted that between 12 and 32 overflow events would occur
per year. The average-22 overflow events per year-is used for comparison with the 4-event-per-
year criterion in the presumption approach. (Note that only one outfall in the system needs to
overflow to trigger the definition of “CSO event” under the presumption approach.)
Once the SWMM model has been hydraulically calibrated, it can be used to predict pollutant
concentrations in the overflow. The summary of the flow-weighted concentrations generated by the
model can then be compared to composite values of actual samples taken during the course of the
overflow. Plots of individual concentrations versus time (pollutographs) can also be used to match
the variation in concentration of a pollutant during the course of the overflow. First flush effects can
also be observed from the model output if buildup/washoff is used.
Model Results
Exhibit 7-12 presents the BOD and total solids output of the SWMM model for the example
storm. Note that the modeled concentrations of both pollutants follow a similar pattern throughout
the overflow event with little if any first flush concentration predicted in the early part of the
overflow. The initial loads assigned within the model for this calibrated example were 70 pounds
per acre for BOD and 1,000 pounds per acre for total solids. This model was previously calibrated
using monitoring data.
Exhibit 7-13 presents predicted and observed values for BOD and total solids concentrations.
The observed concentrations are from analyses of composite samples collected in an automated field
sampler for this storm. The modeled values give an approximate, but not precise, estimate of the
parameters. While some studies have resulted in closer predictions, this discrepancy between
predicted and observed pollutant values is not uncommon.
The modeling in this example could be useful for evaluating the CSS performance against
the four-overflow-event-per-year criterion in the presumption approach. It could also be used to
evaluate the performance of simple controls.
Predicted Observed
BOD TS BOD TS
Flow-weighted concentration (mg/l) 31.4 420 94 300
Example 7-1 is a case study illustrating the CSS and CSO modeling strategy that was
developed and implemented by the City of Indianapolis, Indiana. The City, after carefully evaluating
available options and regulatory requirements, developed this modeling strategy to characterize
system hydraulics and estimate average annual CSO characteristics (i.e., volume, frequency, percent
capture, and pollutant loads). The City used the CSS and CSO models to determine CSO impacts
on the receiving streams (the White River and its tributaries within the City’s combined sewer area),
and is now using the models to evaluate various CSO controls and develop an LTCP.
Recognizing that the interceptor sewers and regulators, not the combined sewers, control wet-
weather system conveyance capacity to the wastewater treatment plants (and therefore control the
occurrences of CSOs), the City used SWMM/EXTRAN to develop a detailed model of interceptor
sewers and regulators that included approximately 82 miles of sewer, 173 regulators, and
134 outfalls. The City used SWMM/RUNOFF to generate runoff flows from drainage
subcatchments and to calibrate wet-weather flow to the EXTRAN model. The City then used the
linked RUNOFF/EXTRAN models to establish critical input data for the STORM model of the CSS,
specifically the regulator/interceptor capacities (STORM “treatment rates”) and the impervious area
estimates (STORM “C” coefficients). The City performed long-term (44-year) continuous
simulations using STORM to compute average annual CSO characteristics. The selected modeling
strategy enabled the City of Indianapolis to accurately determine interceptor sewer conveyance and
system storage capacities, identify system optimization projects, characterize overflows and pollutant
loads to receiving streams, and evaluate various CSO control strategies.
January 1999
CHAPTER 8
This chapter discusses the use of receiving water modeling to evaluate CSO impacts to
receiving waters. It uses the term “modeling” broadly to refer to a range of receiving water
simulation techniques. This chapter introduces simplified techniques, such as dilution and decay
equations, and more complex computer models, such as QUAL2EU and WASP.
Under the CSO Control Policy a permittee should develop a long-term control plan (LTCP)
that provides for attainment of water quality standards (WQS) using either the demonstration
approach or presumption approach. Under the demonstration approach, the permittee documents
that the selected CSO control measures will provide for the attainment of WQS, including designated
uses in the receiving water. Receiving water modeling may be necessary to characterize the impact
of CSOs on receiving water quality and to predict the improvements that would result from different
CSO control measures. The presumption approach does not explicitly call for analysis of receiving
water impacts.
In many cases, CSOs discharge to receiving waters that are water quality-limited and receive
pollutant loadings from other sources, including nonpoint sources and other point sources. The CSO
Control Policy states that the permittee should characterize the impacts of the CSOs and other
pollution sources on the receiving waters and their designated uses (Section II.C.1). Under the
demonstration approach, “[w]here WQS and designated uses are not met in part because of natural
background conditions or pollution sources other than CSOs, a total maximum daily load, including
a wasteload allocation and a load allocation, or other means should be used to apportion pollutant
loads.” (Section II.C.4.b)
Established under Section 303(d) of the CWA, the total maximum daily load (TMDL)
process assesses point and nonpoint pollution sources that together may contribute to a water body’s
impairment. This process relies on receiving water models.
l The type and physical characteristics of the receiving water body. Rivers, estuaries,
coastal areas, and lakes typically require different models.
l The water quality parameters to be modeled. These may include bacteria, DO,
suspended solids, toxics, and nutrients. These parameters are affected by different
processes (e.g., die-off for bacteria, settling for solids, biodegradation for DO, adsorption
for metals) with different time scales (e.g., hours for bacterial die-off, days for
biodegradation) and different kinetics. The time scale in turn affects the distance over
which the receiving water is modeled (e.g., a few hundred feet for bacteria to a few- miles
for DO).
l The number and geographical distribution of CSO outfalls and the need to simulate
sources other than CSOs.
This section discusses some important considerations for hydrodynamic and water quality
modeling of receiving waters, and how these considerations affect the selection and use of a model.
The purpose of receiving water modeling is primarily to predict receiving water quality under
different CSO pollutant loadings and flow conditions in the receiving water. The flow conditions,
or hydrodynamics, of the receiving water are an important factor in determining the effects of CSOs
on receiving water quality. For simple cases, hydrodynamic conditions can be determined from the
receiving water monitoring program; elsewhere a hydrodynamic model may be necessary.
Hydrodynamic and water quality models are either steady-state or transient. Steady-state
models assume that conditions do not change over time, while transient models can simulate
conditions that vary over time. Flexibility exists in the choice of model types; generally, either a
steady-state or transient water quality simulation can be done regardless of whether flow conditions
are steady-state or transient.
A hydrodynamic model provides the flow conditions, characterized by the water depth and
velocity, for which receiving water quality must be predicted. The following factors should be
considered for different water body types:
l
Rivers- Rivers generally flow in one direction (except for localized eddies or other flow
features) and the stream velocity and depth are a function of the flow rate. The flow rate
in relatively large rivers may not increase significantly due to wet weather discharges,
and a constant flow can be used as a first approximation. This constant flow can be a
specified low flow, the flow observed during model calibration surveys, or a flow typical
of a season or month. When the increase of river flow is important, it can be estimated
by adding together all upstream flow inputs or by doing a transient flow simulation. The
degree of refinement required also depends on the time scale of the water quality
parameters of interest. For example, assuming a constant river flow may suffice for
bioaccumulative toxicants (e.g., pesticides) because long-term exposure is ofimportance.
For DO, however, the time variations in river flow rate may be need to be considered.
l Estuaries- CSO impacts in estuaries are affected by tidal variations of velocity and
depth (including reversal of current direction) and by possible salinity stratification.
Tidal fluctuations can be assessed by measuring velocity and depth variations over a tide
cycle or by using a one- or two-dimensional model. Toxics with relatively small mixing
zones can be analyzed using steady currents corresponding to different times during the
tidal cycle, but this may require using a computed circulation pattern from a model.
l Coastal Areas- CSO impacts in coastal areas are also affected by tidal fluctuations. The
discussion on estuaries generally applies to coastal areas, but, because the areas are not
channelized, two-dimensional or even three-dimensional models may be necessary.
l Lakes- CSO impacts in lakes are affected by wind and thermal stratification. Wind-
driven currents can be monitored directly or simulated using a hydrodynamic model
(which may need to cover the entire lake to simulate wind-driven currents properly).
Thermal stratification can generally be measured directly.
1
Because the same basic hydrodynamic equations apply, some of the major models for
receiving waters can be used to simulate more than one type of receiving water body. Ultimately,
three factors dictate whether a model can be used for a particular hydraulic regime. One factor is
whether it provides a one-, two-, or three-dimensional simulation. A second is its ability to handle
specific boundary conditions, such as tidal boundaries.
A third factor is whether the model assumes steady-state conditions or allows for
time-varying pollutant loading. In general, models that assume steady-state conditions cannot
accurately model CSO problems that require analysis of far-field effects. However, in some
instances a steady-load model can estimate the maximum potential effect, particularly in systems
where the transport of constituents is dominated by the main flow of the water body, rather than local
velocity gradients. For example, by assuming a constant source and following the peak discharge
plug of water downstream, the steady-load model QUAL2EU can determine the maximum
downstream effects of conventional pollutants. The result is a compromise that approximates the
expected impact but neglects the moderating effects of longitudinal dispersion. However,
QUAL2EU cannot give an accurate estimate of the duration of excursions above WQS.
The frequency and duration of CSOs are important determinants of receiving water impacts
and need to be considered in determining appropriate time scales for modeling. CSO loads are
1
The basic hydrodynamic equations are for momentum and continuity. The momentum equation describes the
motion of the receiving water, while the continuity equation is a flow balance relationship (i.e., total inflows to the
receiving water less total outflows is equal to the change in receiving water volume).
typically delivered in pulses during storm events. Selection of appropriate time scales for modeling
receiving water impacts resulting from a pulsed CSO loading depends upon the time and space scales
necessary to evaluate the WQS. If analysis requires determining the concentration of a toxic at the
edge of a relatively small mixing zone, a steady-state mixing zone model may be satisfactory. When
using a steady-state mixing zone model in this way, the modeler should apply appropriately
conservative but characteristic assumptions about instream flows during CSO events. For pollutants
such as oxygen demand, which can have impacts lasting several days and extending several miles
downstream of the discharge point, it may be warranted to incorporate the pulsed nature of the
loading. Assuming a constant loading is much simpler (and less costly) to model; however, it is
conservative (i.e., leads to impacts larger than expected). For pollutants such as nutrients where the
response time of the receiving water body may be slow, simulating only the average loading rate,
usually over a period of days (e.g., 21 days) depending on the nutrient, may suffice.
Receiving water models vary from simple estimations to complex software packages. The
choice of model should reflect site conditions. If the pulsed load and receiving water characteristics
are adequately represented, simple estimations may be appropriate for the analysis of CSO impacts.
To demonstrate compliance with the CWA, the permittee may not need to know precisely where in
the receiving water excursions above WQS will occur. Rather, the permittee needs to know the
maximum pollutant concentrations and the likelihood that excursions above the WQS can occur at
any point within the water body. However, since CSOs to sensitive areas are given a higher priority
under the CSO Policy, simulation models for receiving waters with sensitive areas may need to use
short time scales (e.g., hourly pollutant loads), and have high resolution (e.g., several hundred yards
or less) to specifically assess impacts to sensitive areas.
Receiving water models cover a wide variety of physical and chemical situations and, like
combined sewer system (CSS) models, vary in complexity. EPA has produced guidance on
receiving water modeling as part of the Waste Load Allocation (WLA) guidance series. These
models, however, tend to concentrate on continuous sources and thus may not be the most suitable
for CSOs. Ambrose et al. (1988a) summarizes EPA-supported models, including receiving water
models.
This guidance does not provide a complete catalogue of available receiving water models.
Rather, it describes simplified techniques and provides a brief overview of relevant receiving water
models supported by EPA or other government agencies. In many cases, detailed receiving water
simulation may not be necessary. Use of dilution and mixing zone calculations or simulation with
simple spreadsheet models may be sufficient to assess the magnitude of potential impacts or evaluate
the relative merits of various control options.
Types of Simulation
Water quality parameters can be simulated using either single-event, steady-state modeling
or continuous, dynamic modeling. Many systems may find it beneficial to use both types of
modeling.
Many of the simpler approaches to receiving water evaluation assume steady flow and steady
or gradually varying loading. These assumptions may be appropriate if an order-of-magnitude
estimate or an upper bound of the impacts is required. The latter is obtained by using conservative
parameters such as peak loading and low current speed. If WQS attainment is predicted under
realistic worst-case assumptions, more complex simulations may not be needed.
Due to the random nature of CSOs, the use of dynamic simulation may be preferable to
single-event, worst-case, steady-state modeling. Dynamic techniques allow the modeler to derive
the fraction of time during which a concentration was exceeded and water quality was impaired. For
instance, when using daily simulated results, specific concentrations are first ranked with the
corresponding number of occurrences during the simulation period. Frequency distribution plots are
then developed and used to determine how often the l-day-acute water quality criteria are likely to
be exceeded. The same approach can be used to develop frequency distributions for longer periods
such as 4-day or 30-day average concentrations. EPA (1991a) recommends three dynamic modeling
techniques: continuous simulation, Monte Carlo simulation, and lognormal probability modeling.
Monte Carlo simulation is generally used for complex systems that have random
components. Input variables are sampled at random from pre-determined probability distributions
and used in a toxic fate and transport model. The distribution of output variables from repeated
simulations is analyzed statistically to derive a frequency distribution. However, unlike continuous
simulation models, the temporal frequency distribution of the output depends on the temporal
frequency distribution of the input data. For instance, if the water quality criterion is based on a 4-
day average, the input variables must use the probability distributions based on a 4-day average.
The following sections discuss techniques for simulating different water quality parameters
in rivers, lakes and estuaries.
RIVERS
Bacteria and Toxics. Bacteria and toxic contaminants are primarily a concern in the
immediate vicinity of CSO outfalls. They are controlled by lateral mixing, advection, and decay
processes such as die-off (for bacteria), vaporization (for toxics), and settling and resuspension (for
bacteria and toxics). When stream flow is small relative to CSO flow, lateral mixing may occur
rapidly and a one-dimensional model may be appropriate. Initial estimates can be made using a
steady-state approach that neglects the time-varying nature of the CSO. In this case, concentrations
downstream of a CSO are given by:
2 3
where: Cx = max pollutant concentration at distance X from the outfall (M/L )
3
Ce = pollutant concentration in effluent (M/L )
3
Cu = pollutant concentration upstream from discharge (M/L )
3
Qe = effluent flow (L /T)
3
Qu = stream flow upstream of discharge (L /T)
3
Qs = stream flow downstream of discharge, Qu + Qe (L /T)
X = distance from outfall (L)
u = stream flow velocity (L/T)
K = net decay rate (die-off rate for bacteria, settling velocity divided by
stream depth for settling, resuspension velocity divided by stream depth
for resuspension, vaporization rate divided by stream depth for
vaporization) (1/T)
e = 2.71828...
Since bacteria and toxics can settle out of the water column and attach to sediments,
sediments can contain significant amounts of these pollutants. Resuspension of sediments and
subsequent desorption of bacteria and toxics into the water column can be an important source of
receiving water contaminants. Modeling of sediment resuspension requires estimation of
2
M=unit of mass, L=unit of length, and T=unit of time.
resuspension velocities and knowledge of sediment transport processes. Thomann and Mueller
(1987) discusses how to determine the solids balance in a river and estimate sediment resuspension
velocities. Modeling of sediment transport is complex and is often done using computer models
such as WASP5 and HSPF.
In large rivers, lateral mixing may occur over large distances and bacterial counts or toxics
concentrations on the same shore as the discharge can be calculated using the following expression,
as a conservative estimate (U.S. EPA, 1991a):
2
where: Dy = lateral dispersion coefficient (L /T)
W = stream width (L)
= 3.14159...
½
u* = (gds)
2
where: g = acceleration due to gravity (L/T )
s = slope of channel (L/L)
d = water depth (L).
The model DYNTOX (LimnoTech, 1985) is specially designed for analysis of toxics in
rivers and can handle all three dynamic modeling techniques. U.S. EPA (1991a) and the WLA series
by Delos et al. (1984) address the transport of toxics and heavy metals in rivers.
In general, screening analyses using classical steady-state equations can examine DO impacts
to rivers as a result of episodic loads. This approach assumes plug flow, which in turn allows an
assumption of constant loading averaged over the volume of the plug (Freedman and Marr, 1990).
This approach does not consider longitudinal diffusion from the plug, making it a conservative
approach. The plug flow analysis should correlate with the duration of the CSO. For example, a
plug flow simulation of a 2-hour CSO event would result in a downstream DO sag that would also
last for 2 hours. Given the plug flow assumption, the classic Streeter-Phelps equation can estimate
the DO concentration downstream:
This method can address the joint effects of multiple steady sources through the technique
of superposition (Exhibit 8-1). Superposition is used when linear differential equations, such as the
Streeter-Phelps equation, govern pollutant concentrations along a receiving stream. For such linear
systems, the concentration of a pollutant in a river due to multiple steady-state sources is the linear
summation of the responses due to the individual sources. Superposition techniques are also used
to estimate pollutant concentrations due to multiple steady-state sources of toxic pollutants.
However, it cannot address multiple sources that change over time, nor can it address the effects of
river morphology. When such issues are important, more sophisticated modeling techniques are
necessary.
More sophisticated modeling techniques are also necessary to assess the effects of sediment
oxygen demand (SOD) and plant respiration (which remove oxygen from the receiving water), and
photosynthesis by aquatic plants (which adds oxygen to the water). The Streeter-Phelps equation
makes the simplifying assumption that there are only point sources of CBOD, so SOD,
photosynthesis, and respiration are assumed to be zero. If photosynthesis, respiration, and SOD are
significant, more complex analysis is needed to evaluate these factors. These distributed sources and
sinks of DO and BOD are addressed by Thomann and Mueller (1987) and by several computer
models, including QUAL2EU and WASPS.
nutrients and algal growth. Analysis of nutrient impacts in rivers is complex because nutrients and
3
planktonic algae, which are free-floating one-celled algae, usually move through the system rapidly.
The current WLA guidance (U.S. EPA, 1995g) considers only planktonic algae (rather than
all aquatic plants) and discusses nutrient loadings and eutrophication in rivers primarily as a
component in computing DO. The guidance applies to narrative criteria that limit nuisance plant
growth in large, slowly flowing rivers.
LAKES
Bacteria and Toxics. Mixing zone analysis can often be used to assess attainment of WQS
for bacteria and toxics in lakes. For a small lake in which the effluent mixes rapidly, the
concentration response is given by the following equation (Freedman and Marr, 1990):
3
where: C =
concentration (M/L )
M =
mass loading (M)
3
Q =
flow (L /T)
K =
net decay rate (bacteria die-off, settling and resuspension, volatilization,
photolysis, and other chemical reactions) (1/T)
3
V = lake volume (L )
t = time (T).
3
Aquatic plants can be divided into those that move freely with the water (planktonic aquatic plants) and those that
are attached or rooted in place.
and
where the subscripts 1 and 2 refer to the epilimnion (top layer) and hypolimnion (lower layer),
respectively, and variables without subscripts refer to the whole lake, and where:
2
SB = Sediment oxygen demand (M/L -T)
Kd = Deoxygenation coefficient (1/T)
3
L = Steady-state CBOD concentration in water column (M/L ), = W/(Q+KrV),
where W is the mass loading rate, Q is the rate of flow through the lake, V is
the volume, and Kr is the net loss rate.
2
E = Dispersion coefficient (L /T).
Because this analysis assumes steady-state loading and because measuring some of the
parameters proves difficult, the method may only have limited application to CSOs. A modeler able
to define all of the above parameters may choose to apply a more spatially resolved model.
In many cases, complex simulation models are necessary to analyze DO in lakes. These are
either specialized lake models or flexible models, such as EUTROWASP, that are designed to
address issues specific to lakes. Some experienced modelers have been successful in modeling
thermally stratified lakes with one or two dimensional river models (e.g., QUAL2EU) that assume
4
the river bottom is the thermocline.
Nutrient/Eutrophication Impacts. For lakes, simple analytic equations often can analyze
end-of-pipe impacts and whole-lake impacts, but evaluating mixing phenomena frequently requires
a complex computer model (Freedman and Marr, 1990). Simple analytical methods can be applied
to lake nutrient/eutrophication impacts in situations where the CSOs mix across the lake area within
the time scale required to obtain a significant response in the algal population. In most lakes,
phosphorus is considered to be the limiting nutrient for nuisance algal impacts and eutrophication.
Mancini et al. (1983) and Thomann and Mueller (1987) have developed a procedure for calculating
the allowable surface loading rate. The following steps are drawn from this procedure:
Step 1. Estimate the lake volume, surface area, and mean depth.
Step 2. Estimate the mean annual inflow and outflow rates. Where urban areas draining
to the lake constitute a significant fraction of the total drainage area, flow
4
Such techniques should not be used by inexperienced modelers as they can lead to inaccuracies if they are not used
with caution.
estimates from urban runoff and CSOs should be included in the hydrologic
balance around the lake. For lakes with large surface areas, the estimate should
include surface precipitation and evaporation.
Step 3. Determine the average annual total phosphorus loading due to all sources,
including all tributary inflows, municipal and industrial sources, distributed urban
and rural runoff, and atmospheric inputs. Technical Guidance Manual for
Performing Waste Load Allocation (Mancini et al., 1983) discusses techniques
for estimating these loadings.
Step 4. For total phosphorus, assign a net sedimentation loss rate that is consistent with
a local data base.
2
where: is the allowable area1 surface loading rate (M/L -T)
is the trophic state objective concentration of total phosphorus or
3
chlorophyll-a (M/L ),
3
is outflow (L /T),
3
is lake volume (L ),
is mean depth (L), and
is the net sedimentation velocity (L/T).
1
Step 6. Compare the total area1 loading determined in Step 3 to the value of W obtained
in Step 5.
ESTUARIES
Unlike most rivers, estuaries are tidal (i.e., water moves upstream during portions of the tidal
cycle and downstream during other parts of the cycle). When averaged on the basis of tidal cycles,
pollutant transport in narrow, vertically mixed estuaries with dominant longitudinal flow is similar
to that in rivers. However, due to tidal reversals of flow, a narrow estuary may have a much larger
effective dispersion coefficient since shifting tides may cause greater lateral dispersion. In such a
system, the modeler can apply approximate or screening models used for rivers, provided that an
appropriate tidal dispersion coefficient has been calculated. In wider estuaries, tides and winds often
result in complex flow patterns and river-based models would be inappropriate. WLA guidance for
estuaries is provided in several EPA manuals (Ambrose et al., 1990; Martin et al., 1990; Jirka, 1992;
Freedman et al., 1992).
Near-field mixing zone analysis in estuaries also presents special problems, because of the
role of buoyancy differences in mixing. Jirka (1992) discusses mixing-zone modeling for estuaries.
This section describes some computer models relevant to receiving water modeling. Most
of these models are supported by EPA’s Center for Exposure Assessment Modeling (CEAM).
5
CEAM maintains a distribution center for water quality models and related data bases.
CEAM-supported models relevant to modeling impacts on receiving water include QUAL2EU,
WASPS, HSPF, EXAMSII, CORMIX, MINTEQ, and SMPTOX3. The applicability and key
characteristics of the CEAM-supported models are summarized in Exhibit 8-2.
5
See Section 7.3 for information on obtaining models from CEAM.
1
CORMIX was originally developed assuming steady ambient conditions; Version 3 allows for application to some unsteady environments (e.g., tidal reversal
conditions) where transient recirculation and pollutant build-up can occur (CEAM, 1998).
Chapter 8 Receiving Water Modeling
QUAL2EU is a one-dimensional model for rivers. It assumes steady-state flow and loading
but allows simulation of diurnal variations in temperature or algal photosynthesis and respiration.
QUAL2EU simulates temperature, bacteria, BOD, DO, ammonia, nitrate, nitrite, organic nitrogen,
6
phosphate, organic phosphorus, algae, and additional conservative substances. Because it assumes
steady flow and pollutant loading, its applicability to CSOs is limited. QUAL2EU can, however,
use steady loading rates to generate worst-case projections for CSOs to rivers. The model has pre-
and post-processors for performing uncertainty and sensitivity analyses.
Additionally, in certain cases, experienced users may be able to use the model to simulate
non-steady pollutant loadings under steady flow conditions by establishing certain initial conditions
or by dynamically varying climatic conditions. If used in this way, QUAL2EU should be considered
a screening tool since the model was not designed to simulate dynamic quality conditions.
WASP5 is essentially a pollutant fate and transport model. Transport can be driven by
another hydrodynamic model such as DYNHYD5. DYNHYD5 is a one-dimensional/quasi-two-
dimensional model that simulates transient hydrodynamics (including tidal estuaries).
6
A conservative substance is one that does not undergo any chemical or biological transformation or degradation
in a given ecosystem. (U.S. EPA, 1995g)
EXAMSII can rapidly evaluate the fate, transport, and exposure concentrations of steady
discharges of synthetic organic chemicals to aquatic systems. A recent upgrade of the model
considers seasonal variations in transport and time-varying chemical loadings, making it
quasi-dynamic. The user must specify transport fields to the model.
7
CORMIX is an expert system for mixing zone analysis. It can simulate submerged or
surface, buoyant or non-buoyant discharges into stratified or unstratified receiving waters, with
emphasis on the geometry and dilution characteristics of the initial mixing zone. The model uses
a zone approach, in which a flow classification scheme determines which near-field mixing
processes to calculate. The CORMIX model cannot be calibrated in the classic sense since rates are
fixed based on the built-in logic of the expert system.
MINTEQ determines geochemical equilibrium for priority pollutant metals. Not a transport
model, MINTEQ provides a means for modeling metal partitioning in discharges. It provides only
steady-state predictions. The model usually must be run in connection with another fate and
transport model, such as those described above. A number of assumptions (e.g., equilibrium
conditions at the point of mixing between a CSO and the receiving water) must be made to link
MINTEQ predictions to another fate and transport model, so it should be used cautiously in
evaluating wet weather impacts.
7
In some applications CORMIX has proven inaccurate for single port discharges.
toxic substances in the water column and stream bed as a result of point source discharges to streams
and rivers. The model predicts pollutant concentrations in dissolved and particulate phases for the
water column and bed sediments, as well as total suspended solids. SMPTOX3 can be run at three
different levels of complexity: as described above (highest complexity), to calculate toxic water
column concentrations but no interactions with bed sediments (medium complexity), or as a total
pollutant toxics model (low complexity) (LimnoTech, 1992).
8
The following additional models are supported by EPA or other government agencies:
CE-QUAL-W2 is a reservoir and narrow estuary hydrodynamics and water quality model
developed by the Waterways Experiment Station of the U.S. Army Corps of Engineers. The model
provides dynamic two-dimensional (longitudinal and vertical) simulations. It accounts for density
effects on flow as a function of the water temperature, salinity and suspended solids concentration.
CE-QUAL-W2 can simulate up to 21 water quality parameters in addition to temperature, including
one passive tracer (e.g., dye), total dissolved solids, coliform bacteria, inorganic suspended solids,
algal/nutrient/DO dynamics (11 parameters), alkalinity, pH and carbonate species (4 parameters).
8
McKeon and Segna (1987), Ambrose et al. (1988a) and Hinson and Basta (1982) have reviewed some of these
models.
As was the case for CSS models (see Section 7.4), receiving water modeling involves
developing the model, calibrating and validating the model, performing the simulation, and
interpreting the results.
The input data needs for a specific receiving water model depend upon the hydraulic regime
and model used. The permittee should refer to the model’s documentation, the relevant sections of
the WLA guidance, or to texts such as Principles of Surface Water Quality Modeling and Control
(Thomann and Mueller, 1987). Tables B-2 through B-5 in Appendix B contain general tables of data
inputs.
Like CSS models, receiving water models need to be calibrated and validated. The model
should be run to simulate events for which receiving water hydraulic and quality monitoring were
actually conducted, and the model results should be compared to the measurements. Generally,
receiving water models are calibrated and validated first for receiving water hydraulics and then for
water quality. Achieving a high degree of accuracy in calibration can be difficult because:
l Pollutant loading inputs typically are estimates rather than precisely known values.
l Three-dimensional receiving water models are still not commonly used for CSO projects,
so receiving water models involve spatial averaging (over the depth, width or cross-
section). Thus, model results are not directly comparable with measurements, unless the
measurements also have sufficient spacial resolution to allow comparable averaging.
l Loadings from non-CSO sources, such as storm water, upstream boundaries, point
sources, and atmospheric deposition, often are not accurately known.
l Receiving water hydrodynamics are affected by numerous factors which are difficult to
account for. Those include fluctuating winds, large-scale eddies, and density effects.
Although these factors make model calibration challenging, they also underscore the need
for calibration to ensure that the model reasonably reflects receiving water data.
Receiving water modeling can involve single events or long-term simulations. Single event
simulations are usually favored when using complex models, which require more input data and take
significantly longer to run (although advances in computer technology keep pushing the limits of
what can practically be achieved.) Long-term simulations can predict water quality impacts on an
annual basis.
Although a general goal is to predict the number of water quality criteria exceedances,
models can evaluate exceedances using different measures, such as hours of exceedance at beaches
or other critical points, acre-hours of exceedance, and mile-hours of exceedance along a shore.
These provide a more refined measure of the water quality impacts of CSOs and of the expected
effectiveness of different control measures.
CSO loadings commonly are simulated separately from other loadings in order to assess the
relative impacts of CSOs. This is appropriate because the equations that best approximate receiving
water quality are usually linear and so effects are additive (one exception, however, is the non-linear
algal growth response to nutrient loadings).
By calculating averages over space and time, simulation models predict CSO volumes,
pollutant concentrations, and other variables of interest. The extent of this averaging depends on the
model structure, how the model is applied, and the resolution of the input data. The model’s space
and time resolution should match that of the necessary analysis. For instance, the applicable WQS
may be expressed as a 1-hour average concentration not to exceed a given concentration more than
once every three years on average. Spatial averaging may be represented by a concentration
averaged over a receiving water mixing zone, or implicitly by the specification of monitoring
locations to establish compliance with instream criteria. In any case, the permittee should note
whether the model predictions use the same averaging scales required in the permit or relevant WQS.
When used for continuous rather than event simulation, as suggested by the CSO Control
Policy, simulation models can predict the frequency of exceedances of water quality criteria.
Probabilistic models, such as the Monte Carlo simulation, also can make such predictions. In
probabilistic models, the simulation is made over the probability distribution of precipitation and
other forcing functions such as temperature, point sources, and flow. In either case, modelers can
analyze the output for the frequency of water quality criteria exceedances.
The key result of receiving water modeling is the prediction of future conditions due to
implementation of CSO control alternatives. In most cases, CSO control decisions will have to be
supported by model predictions of the pollutant load reductions necessary to achieve WQS. In the
receiving waters, critical or design water quality conditions might be periods of low flows and high
temperature that are established based on a review of available data. Flow, temperature, and other
variables for these periods then form the basis for analysis of future conditions.
This chapter focuses on the link between CSOs and the attainment of water quality standards
(WQS). As discussed in previous chapters, permittees can consider a variety of methods to analyze
the performance of the combined sewer system (CSS) and the response of a water body to pollutant
loads. Permittees can use these methods to estimate the water quality impacts of a proposed CSO
control program and evaluate whether it is adequate to meet CWA requirements.
Under the CSO Control Policy, permittees need to develop long-term control plans (LTCPs)
that provide for WQS attainment using either the presumption approach or the demonstration
approach. This chapter focuses primarily on issues related to the demonstration approach since this
approach requires the permittee to demonstrate that the selected CSO controls will provide for the
attainment of WQS. As mentioned in Chapter 8, the presumption approach does not explicitly call
for analysis of receiving water impacts and thus generally involves less complex modeling.
Modeling time-varying wet weather sources such as CSOs is more complex than modeling
more traditional point sources. Typically, point-source modeling assumes constant pollutant loading
to a receiving water body under critical, steady-state conditions-such as the minimum seven-
consecutive-day average stream flow occurring once every ten years (i.e., 7Q10). Wet weather loads
occur in pulses, however, and often have their peak impacts under conditions other than low-flow
situations. This makes modeling the in-stream impact of CSOs more complicated than modeling
the impacts of steady-state point source discharges such as POTWs. A receiving water model must
therefore accommodate the short-term variability of pollutant concentrations and flow volume in the
discharge as well as the dynamic conditions in the receiving water body. Notwithstanding these
limitations, however, properly-applied modeling techniques can be useful in analyzing the impact
of CSOs on receiving waters.
CSO pollutant loads can be incorporated into receiving water models using either a steady-
state or a dynamic approach, as discussed in Chapter 8. A steady-state model can provide an
approximate solution using, for example, average loads for a design storm. A dynamic approach
incorporates time-varying loads and simulates the time-varying response of the water body. The
steady-state approximation uses some average conditions that do not account for the time-varying
nature of flows and loads. Thus a steady-state model may provide less exact results, but typically
requires less cost and effort. A dynamic model requires more resources but may result in a more
cost-effective CSO control plan, since it does not use some of these simplifying assumptions.
Generally, the modeler should use the simplest approach that is appropriate for local
conditions. A steady-state model may be appropriate in a receiving water that is relatively
insensitive to short-term variations in load rate. For instance, the response time of lakes and coastal
embayments to some pollutant loadings may be measured in weeks to years, and the response time
of large rivers to oxygen demand may be measured in days (Donigian and Huber, 1991). Steady-
state models are also useful for estimating the dilution of pollutants, such as acute toxins or bacteria,
close to the point of release.
The demonstration approach requires the permittee to show that its selected CSO controls
will provide for attainment of WQS. The CSO Control Policy states that:
i. the planned control program is adequate to meet WQS and protect designated uses,
unless WQS or uses cannot be met as a result of natural background conditions or
pollution sources other than CSOs;
ii. the CSO discharges remaining after implementation of the planned control program
will not preclude the attainment of WQS or the receiving waters’ designated uses or
contribute to their impairment. Where WQS and designated uses are not met in part
because of natural background conditions or pollution sources other than CSOs, a total
maximum daily load, including a wasteload allocation and a load allocation, or other
means should be used to apportion pollutant loads... (Section II.C.4.b)
The first step in analyzing CSO impacts on receiving water is to identify the pollutants or
stressors of concern and the corresponding WQS. CSOs are distinguished from storm water loadings
by the increased levels of such pollutants as bacteria, oxygen-demanding wastes, and certain
nutrients. In some cases, toxic pollutants entering the CSS from industrial sources also may be of
concern.
State WQS include designated uses and both numerical and narrative water quality criteria.
Since CSO controls must ultimately provide for attainment of WQS, the analysis of CSO control
alternatives should be tailored to the applicable WQS. For example, if the water quality criterion
of concern is expressed as a daily average concentration, the analysis should address daily averages.
Many water bodies have narrative criteria such as a requirement to limit nutrient loads to an amount
that does not produce a “nuisance” growth of algae, or a requirement to prevent solids and floatables
build-up. In such cases, the permittee could consider developing a site-specific, interim numeric
performance standard that would result in attainment of the narrative criterion.
As noted in Chapter 2, a key principle of the CSO Control Policy is the review and revision,
as appropriate, of WQS and their implementation procedures. In identifying applicable WQS, the
permittee and the permitting and WQS authorities should consider whether revisions to WQS are
appropriate for wet weather conditions in the receiving water.
EPA’s water quality criteria assist States in developing numerical standards and interpreting
narrative standards (U.S. EPA, 1991a). EPA recommends that water quality criteria for protection
of aquatic life have a magnitude-duration-frequency format, which requires that the concentration
of a given constituent not exceed a critical value more than once in a given return period:
l Duration- The averaging period, which is the period of time over which the in-stream
concentration is averaged for comparison with criteria concentrations. This specification
limits the duration of concentrations above the criteria.
Many States rely instead on the concept of design flows, such as 7Q10. Evaluating
compliance at a design low flow of specified recurrence is a simple way to approximate the average
duration and frequency of excursions above the WQS. A single critical low flow, however, is not
necessarily the best choice for wet-weather flows, which may not occur simultaneously with drought
conditions. Consequently, a design flow-based control strategy may be overly conservative, and
suitable mainly for situations where monitoring data are very limited or areas are highly sensitive.
Some water quality criteria are expressed in formats that vary from the magnitude-duration-
frequency format. In some cases, such as State WQS for indicator bacteria, water quality criteria are
expressed as an instantaneous maximum and a long-term average component. The long-term
average component of water quality criteria for fecal coliforms typically specifies a 30-day
geometric mean or median, and a certain small percentage of tests performed within a 30-day period
that may exceed a particular upper value. For dissolved oxygen (DO) and pH, State criteria may be
expressed as fixed minimum concentrations, rather than as magnitude-duration-frequency.
The statistical form of the relevant WQS is important in determining an appropriate model
framework. Does the permittee need to calculate a long-term average, a worst case maximum, or
an actual time sequence of the number of water quality excursions? An approach that gives a
reasonable estimate of the average may not prove useful for estimating an upper bound.
Receiving water impacts can be analyzed at varying levels of complexity, but all approaches
attempt to answer the same question: Using a prediction of the frequency and volume of CSO events
and the pollutant loads delivered by these events, can WQS in the receiving water body be attained
with a reasonable level of assurance?
Any of the following types of analyses, arranged in order of increasing complexity, can be
used to answer this question:
l Design Flow Analysis- This approach analyzes the impacts of CSOs under the
assumption that they occur at a design condition (e.g., 7Q10 low flow prior to addition
of the CSO flow). The CSO is added as a steady-state load. If WQS can be attained
under such a design condition, with the CSO treated as a steady source, WQS are likely
to be attained for the actual wet weather conditions. This approach is conservative in two
respects: (1) it does not account for the short-term pulsed nature of CSOs, and (2) it does
not account for increased receiving water flow during wet weather.
l Design Flow Frequency Analysis- Where the WQS is expressed in terms of frequency
and duration, the frequency of occurrence of CSOs can be included in the analysis. The
design flow approach can then be refined by determining critical design conditions that
can reasonably be expected to take place concurrently with CSOs. For instance, if CSO
events occur primarily in one season, the analysis can include critical flows and other
conditions appropriate to that season, rather than the 7Q10.
l Statistical Analysis- Whereas the previous two approaches rely on conservative design
conditions, a statistical analysis can be used to consider the range of flows that may occur
together with CSO events. This analysis more accurately reflects the frequency of WQS
excursions.
l Watershed Simulation- A statistical analysis does not consider the dynamic relationship
between CSOs and receiving water flows. For example, both the CSO and receiving
water flows increase during wet weather. Demonstrating the availability of this
additional capacity, however, requires a model that includes the responses of both the
sewershed and its receiving water to the rainfall events. Dynamic watershed simulations
may be carried out for single storm events or continuously for multiple storm events.
The permittee should consider the tradeoffs between simpler and more complex types of
receiving water analysis. A more complex approach, although more costly, can generally provide
more precise analysis using less conservative assumptions. This may result in a more tailored, cost-
effective CSO control strategy.
This section presents three examples to illustrate key points for analyzing CSO impacts on
receiving waters. The examples focus on (1) establishing the link between model results and
demonstrating the attainment of WQS, and (2) the uses of receiving water models at different levels
of complexity, from design flow analysis to dynamic continuous simulation.
The first example shows how design flow analysis or more sophisticated methods can be
used to analyze bacteria loads to a river from a single CSO event. The second example, which is
more complex, involves bacterial loads to an estuary. The third example illustrates how biochemical
oxygen demand (BOD) loads from a CSS contribute to DO depletion.
This example involves a CSS in a small northeastern city that overflows relatively frequently
and contributes to WQS excursions. CSOs are the only pollutant source, and only a single water
quality criterion--for fecal coliforn-applies. The use classification for this receiving water body
is primary and secondary contact recreation. The city has planned several engineering improvements
to its CSS and wishes to assess the water quality impacts of those improvements.
Boat Launch
Chapter 9 Assessing Receiving Water Impacts and Attainment of WQS
In this example, dilution calculations may suffice to predict whether the water quality
criterion is likely to be attained during a given CSO event. This is because:
(1) The State allows mixing zones, so the water quality criterion must be met at the edge of
the mixing zone. If the criterion is met there, it will also be met at points farther away.
(2) Die-off will reduce the numbers of bacteria as distance from the discharge increases.
(3) Since the river flows constantly in one direction, bacterial concentrations do not
accumulate or combine loads from several days of release.
To illustrate the various levels of receiving water analysis, this example assumes that the
magnitude and timing of CSOs can be predicted precisely and that the long-term average
characteristics of the CSS will remain constant. In the absence of additional CSO controls, the
predictions for the next 31 years include the following (Exhibit 9-2):
1
(1) The system should experience a total of 238 overflow events, an average of 7.7 per year.
(2) The largest discharge is approximately 1.1 million cubic feet, but most of the CSOs are
less than 200,000 cubic feet.
(3) The maximum number of overflow events in any one month is 18.
(4) During that month, the maximum receiving water concentration resulting from CSOs
exceeds 6,000 MPN/100 ml. Even in this “worst-case” month, however, the geometric
mean is 400 MPN/100 ml, based on 30 daily samples and assuming a background
concentration of 100.
At least one CSO event occurs in each calendar month, although 69 percent of the events
occur in March and April when snowmelt increases flow in the CSS. Because river flow is lower
in summer and fall, the rarer summer and fall CSOs may cause greater impact in the receiving water.
1
An overflow event is the discharge from one or more CSO outfalls as the result of a single wet weather event. In
this example, the number and volume of CSOs pertains to the discharges from the single outfall.
The applicable water quality criterion for fecal coliforms specifies that:
(1) The geometric mean for any 30-day period not exceed 400 MPN (“most probable
number”) per 100 ml, and
(2) Not more than 10 percent of samples taken during any 30-day period exceed 1,000 MPN
2
per 100 ml.
The water quality criterion does not specify an instantaneous maximum count for this use
classification.
In general, the second component of the water quality criterion-a percentile (or maximum)
standard-will prove more restrictive for CSOs. A CSS that overflows less than 10 percent of the
time (fewer than 3 days per month) could be expected to meet a not-more-than-10-percent
requirement, on average, but probably only if loads from other sources were well below
1000 MPN/100 ml and the CSS discharged to a flowing river system, where bacteria do not
accumulate from day to day. It is possible that an actual overflow event might not result in an
excursion above the 1000 MPN/100 ml criterion if the flow in the receiving water were sufficiently
large. The permittee, however, must demonstrate that the likelihood of a 30-day period when CSOs
result in non-attainment of the WQS more than 10 percent of the time is extremely low. This means
that the analysis must consider both the likelihood of occurrence of overflow events and the dilution
2
Most Probable Number (MPN) of organisms present is an estimate of the average density of fecal coliforms in a
sample, based on certain probability formulas.
3
The geometric mean, which is defined as the antilog of the average of the logs of the data, typically approximates
the median or midpoint of the data.
capacity of the receiving water at the time of an overflow. The following sections demonstrate
various ways to make this determination.
The CSO outfall in this example is at a bend in the river where mixing is rapid. Therefore,
the loads are considered fully mixed through the cross-section of flow. The concentration in the
receiving water is determined by a simple mass balance equation,
where C represents concentration and Q flow (in any consistent units). The subscripts RW, CSO,
and U refer to “receiving water, ” “combined sewer overflow,” and “upstream,” respectively.
For the design flow analysis, upstream volume Qu is set to a low flow of specified recurrence
and receiving water concentration CRW is set equal to the water quality criterion. In this example,
upstream volume QU is set at the 7Q10 flow. The 7Q10 flow is commonly used for steady-state
wasteload analyses; although it has a lo-year recurrence and is much more stringent than the
not-more-than-10-percent requirement of the standard, this conservatism ensures that excursions of
the standard will indeed occur only rarely.
The 7Q10 flow in this river is 313.3 cfs, so upstream volume QU, is set to 313.3. The
background (upstream) fecal coliform concentration is 100 MPN/100ml, so CU is set to 100. The
WQS stipulates that not more than 10 percent of samples taken during any 30-day period exceed
1,000 MPN/100 ml; thus receiving water concentration CRW is set at 1000. Given 7Q10 flow in the
receiving water, the mass balance equation may be rearranged to express the CSO concentration that
just meets the standard, in terms of the CSO flow volume:
The equation treats both the concentration and flow from the CSO as variables, unlike a standard
wasteload allocation for a point source, where flow is usually considered constant. For a given CSO
concentration, the capacity of the receiving water increases as increased CSO volume provides
additional dilution capacity. Therefore, the relationship between allowable concentration and CSO
flow is not linear. The necessary levels of control on CSOs are not represented by a single point,
but rather by a set of combinations of concentration and flow that meet the water quality criterion.
Exhibit 9-3 shows combinations of CSO concentration and CSO flow that just meet the WQS
at 7Q10 flow. The region below the line represents potential control strategies. For instance, for
CSO flows below 1 cfs, the WQS would be met at the design low flow of 313.3 cfs in the receiving
6
water when the concentration in the CSO remained below 0.28 x 10 MPN/100 ml. At a CSO flow
of 6 cfs, however, the concentration must be below 0.048 x 106 MPN/100 ml for WQS to be
attained.
6
Since the typical concentration of fecal coliforms in CSOs is approximately 2 x 10
MPN/100 ml, demonstrating attainment of the water quality criterion via a design low flow analysis
would be difficult.
A design low flow analysis is often conservative because CSOs typically occur when the
receiving water is responding to precipitation and higher-than-normal dilution capability is available.
Further, while CSOs may occur during design low flows, this will be much rarer than the occurrence
of the low flows themselves. Therefore, the use of the design low flow protects to a more stringent
level than indicated since dilution effects are likely to be greater. Dilution effects can be
considerable in areas of multiple sources of storm water discharge. Design flow analysis is usually
not sufficient in circumstances involving multiple storm water discharges, highly sensitive habitats,
and river areas particularly prone to sediment deposition.
Since most CSOs occur in spring, the probability of a water quality criterion exceedance
needs to be calculated on a month-by-month rather than annual average basis. Here, reducing the
relatively high number of overflows in April should result in attainment of the criterion in other
months.
Additional refinements can focus more specifically on eliminating only those CSO events
predicted to exceed WQS at actual receiving water flow. Not all of the April events result in such
excursions; many are very small. Further, the dilution capacity of the receiving water tends to be
high during the spring. Therefore, the analysis can be refined by considering a design flow
appropriate to the month in question and then counting only those CSO events predicted to result
in excursions above WQS at this flow. The resulting table of predicted receiving water
concentrations can be analyzed to determine the percentage reduction in CSO volume needed to meet
the WQS.
The design flow frequency analysis can give results that are overly conservative, because the
analysis assumes low flow at the same time that it imposes a low probability of exceeding the
standard at that low flow. This approach, then, pays a price for its simplicity, by requiring highly
conservative assumptions. A less restrictive analysis would need information on the probability
distribution of receiving water flows likely to occur during CSO events.
Statistical Analysis
The next level considers not only design low flows, but the whole range of flows experienced
during a month. Although CSOs are more likely when receiving water flow is high, CSO events do
not always have increased dilution capacity available. Clearly, however, CSOs will experience at
least the typical range of dilution capacities. Therefore, holding the probability of excursions to a
specified low frequency entails analyzing the impacts of CSOs across the possible range of receiving
water flows, and not only design low flows.
This example assumes that the permittee has a predictive model of CSO volumes and
concentrations and adequate knowledge of the expected distribution of flows based on 20 or more
years of daily gage data. In short, the permittee knows the loads and the range of available dilution
capacity but not the frequency with which a particular load will correspond to a particular dilution
capacity. A Monte Carlo simulation can readily address this type of problem, and is used with data
on CSOs in April, since this is the month with the highest average number of CSOs and is the only
4
month in which overflows occur more than 10 percent of the time, on average.
Exhibit 9-4 summarizes the April receiving water flows in a flow-duration curve, which
indicates the percent of time a given flow is exceeded. The distribution of flows is asymmetrical,
with a few large outliers. An analysis of flow data indicates that daily flows typically are
lognormally distributed. April’s flows are lognormal with mean natural log of 7.09, which is
5
ln (1,200 cfs) , and standard deviation of 0.46.
The 31 years of CSS data include 96 overflow events in April. In the Monte Carlo simulation
these 96 events were matched with randomly selected receiving water flows from the April flow
distribution, for a total of 342 “Aprils” of simulated data. The number of events in which the
1,000 MPN/100 ml standard would be exceeded was then calculated, and the count for the month
tabulated.
Exhibit 9-5 shows the results. Of the 342 Aprils simulated, 122 had zero excursions of the
standard attributable to the CSS. The maximum number of predicted excursions in any April was
17. The average number for the month was 2.45.
This analysis more closely approaches the actual pattern of water quality excursions caused
by the CSS. The objective implied by the WQS is three or fewer excursions per month. In
Exhibit 9-5, the right-hand axis gives the cumulative frequency of excursions, expressed on a
4
The Monte Carlo approach describes statistically the components of the calculation procedure or model that are
subject to uncertainty. The model (in this case, the simple dilution calculation) is run repeatedly, and each time the
uncertain parameter, such as the receiving water flow, is randomly drawn from an appropriate statistical distribution.
As more and more random trials are run, the resulting predictions build up an empirical approximation of the distribution
of receiving water concentrations that would result if the CSO series were repeated over a very long series of natural
flows. Monte Carlo analysis can often be performed using a spreadsheet. The resulting distribution can then be used
for analyzing control strategies. Also see discussion in Section 8.3.
5
For a lognormal distribution, the mean is equal to the natural log of the median of the data (7.09 = ln (median)).
Therefore, the median April flow = e7.09 = 1,200 cfs.
12000
10000
8000
6000
4000
2000
0
0 20 40 60 80 100
Percent of Time Exceeded
Chapter 9 Assessing Receiving Water Impacts and Attainment of WQS
zero-to-one scale. Of the 342 simulated Aprils, over 75 percent were predicted to have three or
fewer excursions, leaving 25 percent predicted to have four or more. Note that the 11 simulated
Aprils with either 16 or 17 excursions all result from the same month of CSS data, corresponding
to an abnormally wet period.
Once set up, the Monte Carlo simulation readily evaluates potential control strategies. For
instance, to evaluate a control strategy with the goal of a 20-percent reduction in CSO flow and a
30-percent reduction in coliform levels, the Monte Carlo simulation is rerun for these reduced CSO
flows and coliform levels. The results show that of the 342 simulated Aprils, 82 percent were
predicted to meet the water quality criterion. Although the Monte Carlo analysis introduces a
realistic distribution of flows, it may still result in an overly conservative analysis for how CSOs
correlate with receiving water flows, since it involves using a distribution, such as lognormal, which
6
at best approximates the true distribution of flows. A more exact analysis needs accurate
information about the relationship between CSO flows and loads and receiving water dilution
capacity.
6
An analysis of flow distribution must be made so that the appropriate Monte Carlo distribution and range are
calculated.
analysis of the data, introduces complexity. This analysis uses a model that accurately predicts the
available dilution capacity corresponding to each CSO event. Such a model accurately represents
the actual coliform counts in the receiving water and enables the permittee to determine which events
exceed the standard of 1,000 MPN/100 ml.
Exhibit 9-6 presents the results as the count of CSO events by month which result in
receiving water concentrations greater than or equal to 1,000 MPN/100 ml. For 31 years of data,
only three individual months are predicted to have more than three days (i.e., greater than 10 percent
of the days in a month) in excess of the standard. Consequently, excursions above the monthly
percentile goal occur only about 0.8 percent of the time. Further, the return period for years with
exceedances of this standard is 10.3 years (3 occurrences over 31 years). Although the CSS
produces relatively frequent overflows, the rate of actual WQS exceedances is quite low.
Exhibit 9-7, which plots CSO volumes versus receiving water flow volume, illustrates why WQS
exceedances remain rare. This figure shows that all the CSO events have occurred when the
receiving water is at flow above 7Q10. Furthermore, most of the large CSO discharges are
associated with receiving water flows well above low flow. Although this excess dilution capacity
reduces the effect of the CSO pollutant loads, demonstrating compliance also necessitates careful
documentation of the degree of correlation.
The second example involves bacterial WQS in a tidal estuary. Like the previous example,
it attempts to estimate the frequency of excursions of the WQS. However, the fate and transport of
bacteria in an estuarine system is more complex than the transport in freshwater systems. Estuaries
are both dispersive and advective in nature which creates considerable variations in the water quality.
Dispersion is caused by the effects of tidal motion, which is the result of upstream and downstream
currents. Advection is the result of the freshwater flow-through in the estuary. Exhibit 9-8 is a map
of the estuary with the locations of the CSO outfall, mixing zone, and two sensitive areas (beach and
shellfish bed) with more-restrictive bacterial standards.
As in the previous example, WQS for fecal coliform are expressed as a geometric mean of
400 MPN/100 ml and not more than 10 percent of samples in a 30-day period above
1,000 MPN/100 ml. The shell fishing and bathing areas have more restrictive WQS, specifying that
the 30-day geometric mean of fecal coliform counts not exceed 200 MPN/100 ml on a minimum of
five samples and that no more than 20 percent of samples exceed 400 MPN/100 ml.
1. The estuary is one-dimensional. It is not strongly stratified near the source and the
longitudinal gradient of bacterial concentration is dominant.
3. The estuary is in a steady-state condition and area, flow, and reaction rate are constant
with distance.
Estuary
Chapter 9 Assessing Receiving Water Impacts and Attainment of WQS
Under these assumptions, the following mass balance equation can be derived for an infinitely long
estuary with a waste input at x = 0. This differential equation is often referred to as the one-
dimensional advection-dispersion equation.
for n = n0 at x= 0 (2)
n=0 at x= +/- (3)
2
where E is the tidal dispersion (mi /day), U = Q/A the net non-tidal velocity, K is the bacteria die-off
rate (/day), and n is the bacterial concentration (MPN/100 ml).
The solutions to equation (1) with conditions (2) and (3) are:
n = n0 exp(j1x) for x 0
n = n0 exp(j2x) for x 0
2
The ratio KE/U , referred to as the Estuary Number, strongly controls the character of the
2
solution. As KE/U approaches zero, advection predominates and the concentrations in the estuary
2
become increasingly similar to the transport in a stream and, as KE/U becomes large, the
concentrations approach those in a purely dispersive system. Note that in a well-mixed river with
no tides, a is equal to 1, and n0 is given by the input CSO load divided by the flow. In an estuary,
the concentration is reduced by the coefficient a due to the transport of the substance upstream and
downstream because of tidal mixing.
Exhibit 9-9 displays the results of this analysis. It predicts fecal coliform counts at different
locations in the estuary under different assumptions for tidal dispersion and non-tidal velocity.
It is most appropriate to compare the geometric mean criteria to the 30Q10 upstream flow
and average load (since the standard is written as a 30-day average), and the percentile standards to
the 7Q10 upstream flow and event maximum load. Scoping indicates that the CSOs may cause the
short-term criterion to be exceeded at the mixing zone boundaries and may cause impairment at the
up-estuary beach. Increasing the estimate of the dispersion coefficient increases the estimated
concentration at the beach, reflecting increased up-estuary “smearing” of the contaminant plume,
which illustrates that the minimum mixing power may not be a reasonable design condition for
evaluating maximum impacts at points away from the outfall. Potential WQS excursions at the
beach are a concern only at low upstream flows, since the combination of average loads and 30Q10
freshwater flows is not predicted to cause impairment. In evaluating impacts at the beach, recall that
scoping was conducted using a one-dimensional model, which averages a cross-section. If the
average is correctly estimated, impacts at a specific point (e.g., the beach) may still differ from the
average. Concentrations at the beach may be higher or lower than the cross-sectional average,
depending on tidal circulation patterns.
Statistical Analysis
The design flow analyses of the previous two sections contain a number of conservative
simplifying assumptions:
(2) They assume a design minimum dilution capability for the estuary
(3) They do not account for many of the real-world complexities of estuarine mixing
(4) They do not account for the effects of temperature and salinity on bacterial die-off.
The scoping analysis can be improved by considering a full distribution of probable upstream
flows in a Monte Carlo simulation. The expected range of hydrodynamic dispersion coefficients
could also be incorporated into the analysis.
Watershed Simulation
Building a realistic model of contaminant distribution and transport in estuaries is typically
resource-intensive and demanding. A watershed simulation may, however, be needed to demonstrate
compliance for some systems where the results of conservative design flow analyses are unclear.
Detailed guidance on the selection and use of estuarine models is provided in EPA’s Wasteload
Allocation series, Book III (Ambrose et al, 1990; Martin et al., 1990).
The third example concerns BOD and depletion of DO, another important water quality
concern for many CSSs. Unlike bacterial loads, BOD impacts are usually highest downstream of
the discharge and occur some time after the discharge has occurred.
The CSS in an older industrial city has experienced frequent overflow events. The CSOs
discharge to a moderate-sized river on a coastal plain. In the reach below the CSS discharge, the
river’s 7Q10 flow is 194 cfs, with a depth of 5 feet and a velocity of 0.17 ft/s. Above the city,
velocities range from 0.2 to 0.3 ft/s at 7Q10 flow. A major industrial point source of BOD lies
18 miles upstream. A POTW with advanced secondary treatment discharges three miles upstream
of the CSO (Box 9-3).
The river reach below the city has a designated use of supporting a warm water fishery. For
this designation, State criteria for DO are a 30-day mean of 7.0 mg/l and a l-day minimum of
5.0 mg/l. The State also requires that WLAs for BOD be calculated on the basis of the l-day
minimum DO standard calculated at 7Q10 flow and the maximum average monthly temperature.
The 5.0 mg/l criterion is not expressed in a frequency-duration format; the 1-day minimum is a fixed
value, but evaluation in terms of an extreme low flow of specified recurrence implicitly assigns an
7
Similar DO analysis is discussed in Thomann and Mueller (1987).
was assumed for other reaches upstream of the CSO. This is a simplifying assumption that is
sufficient for the scoping analysis described here. SOD in the river reach below the CSO has been
included in the analysis since this is the reach of concern. Since there are many sources of SOD
other than CSOs, contributions of SOD from other sources should be considered at the next level of
analysis.
Results of the scoping model application are shown in Exhibit 9-10, which shows the
interaction of the point source, POTW, and CSO. The exhibit combines two worst-case conditions:
high flow from the episodic source and low (7410) flow in the receiving water. Under these
conditions, the maximum DO deficit is expected to occur 7.5 miles downstream of the CSO, with
predicted DO concentrations as low as 3.9 mg/l. Under such conditions, the CSO flow is
approximately 25 percent of total flow in the river.
To accommodate this relationship, the design flow model can be modified to assess the
dependence of DO concentrations on upstream flow during maximum likely loading from the CSO.
Design flow was simulated using the worst-case CSO flow over a variety of concurrent upstream
8
The average frequency of excursions is intended to provide an average period of time during which aquatic
communities recover from the effects of the excursion and function normally before another excursion. Based on case
studies, a three-year return interval was determined to be appropriate. The three-year return interval was linked to the
7Q10 flow since this flow is generally used as a critical low flow condition.
River Mile
Chapter 9 Assessing Receiving Water Impacts and Attainment of WQS
flows, since upstream flows affect both the dilution capacity of the river and the velocity of flow and
reaeration rate. As shown in Exhibit 9-11, the estimated DO concentrations depend strongly on
upstream flow. Note that WQS are predicted to be attained if the upstream flow is greater than about
510 cfs. A flow less than 510 cfs occurs about five times per year, on average, in this segment of
the river.
The target rate of WQS excursions is one in three years. An upper bound for the actual
long-term average rate of excursions can be established as the probability that flow is less than
510 cfs in the river multiplied by the probability that a CSO occurs:
where Pexc is the probability of a WQS excursion on any given day and fcso is the fraction of days in
the year on which CSO discharges occur, on average. Since the goal for excursions is once every
three years, Pexc is set at 1/(3 x 365), or .000913. Since a flow less than 510 cfs occurs five times
per year, p(Q<510) is 5/365, or .0137. Substituting these values into the equation yields
fcso = .000913/.0137 = 0.067. This implies that up to 24 CSOs per year will meet the long-term
average goal for DO WQS excursions, even under the highly conservative assumption that all CSOs
provide the reasonable maximum BOD load.
An important caveat, however, is that no other significant wet weather sources are assumed
to be present in the river. In most real rivers, major precipitation events also produce BOD loads
from storm water, agriculture, etc. Where such loads are present, conservative assumptions
regarding these additional sources need to be incorporated into the scoping level frequency analysis.
January 1999
Chapter 9 Assessing Receiving Water Impacts and Attainment of WQS
As with the other examples, further refinement in the analysis can be attained by examining
the statistical behavior of the CSO and receiving water flows in more detail. For example, the use
of a constant CSO load is a conservative, simplifying assumption that is appropriate for the scoping
level analysis presented here. Dynamic continuous simulation models could be used to provide a
more realistic estimate of the actual time series of DO concentrations resulting from CSOs.
9.4 SUMMARY
As illustrated in the preceding examples, no one method is appropriate for a particular CSS
or for all CSSs, and a complex dynamic simulation is not always necessary. The method should be
appropriate for the receiving water problem. The municipality (in cooperation with the NPDES
authority) needs to balance effort spent in analysis with the level of accuracy required. However,
as the first example illustrated, as additional effort is invested assumptions can usually be refined
to better reflect the actual situation.
Alley, W.M. and P.E. Smith. 1982a. Distributed Routing Rainfall-Runoff Model, Version IL
Open-File Report 82-344. USGS, Denver, CO.
Alley, W.M. and P.E. Smith. 1982b. Multi-Event Urban Runoff Quality Model. Open-File
Report 82-764. USGS, Denver, CO
Alley, W.M. 1977. Guide for Collection, Analysis, and Use of Urban Storm Water Data: A
Conference Report. Cosponsored by the Engineering Foundation, U.S. Geological
Survey, and the American Society of Civil Engineers. ASCE, New York, NY.
Ambrose, R.B., J.L. Martin and J.F. Paul. 1990. Technical Guidance Manual for Performing
Waste Load Allocations, Book III: Estuaries, Part 1, Estuaries and Waste Load
Allocation Models. EPA 823/R-92-002. U.S. EPA Office of Water.
Ambrose, R.B., J.P. Connolly, E. Southerland, T.O. Barnwell and J.L. Schnoor. 1988a. “Waste
allocation simulation models.” Journal WPCF. 60: 1646-1655.
Ambrose, R.B., T.A. Wool, J.P. Connolly and R.W. Schanz. 1988b. WASP4, A Hydrodynamic
and Water Quality Model - Model Theory, User's Manual, and Programmer‘s Guide.
EPA 600/3-87/039. Environmental Research Laboratory, Athens, GA.
American Public Health Association (APHA). 1992. Standard Methods for the Analysis of
th
Water and Wastewater, 18 Edition. Washington, DC.
American Society for Testing and Materials (ASTM). 1991. Standard Guide for Conducting
Sediment Toxicity Tests with Freshwater Invertebrates. ASTM E-1383-94. Philadelphia,
PA.
Association of Metropolitan Sewerage Agencies (AMSA). 1996. Performance Measures for the
National CSO Control Program. Washington, DC.
Bedient, P.B. and W.C. Huber. 1992. Hydrology and Floodplain Analysis. Second Edition.
Addison-Wesley Publishing Company, New York, NY.
Bowie, G.L. et al. 1985. Rates, Constants, and Kinetic Formulations in Surface Water Quality
Modeling (2d Edition). EPA 600/3-85/040. Environmental Research Laboratory,
Athens, GA
Brown, D.S. and J.D. Allison. 1987. MINTEQA1, An Equilibrium Metal Speciation Model:
User’s Manual. EPA 600/3-87/012.
Brown, L.C. and T.O. Barnwell. 1987. The Enhanced Stream Water Quality Model QUAL2E
and QUAL2E-UNCAS: Documentation and User’s Manual. EPA 600/3-87/007.
Crowder, L.B. 1990. “Community Ecology.” in C.B. Schreck and P.B. Moyle (editors). 1990.
Methods for Fish Biology. American Fisheries Society, Bethesda, MD. pp 609-632.
Delos, C.G. et al. 1984. Technical Guidance Manual for Performing Waste Load Allocations,
Book II: Streams and Rivers, Chapter 3: Toxic Substances. EPA 440/4-84-022.
Dennis, W.M. and B.G. Isom (editors). 1984. Ecological Assessment of Macrophyton:
Collection, Use, and Meaning of Data. ASTM Special Technical Publication 843.
American Society for Testing and Materials, Philadelphia, PA.
Doneker, R.L. and G.H. Jirka. 1990. Expert System for Hydrodynamic Mixing Zone Analysis of
Conventional and Toxic Submerged Single Port Discharges (CORMIX1). EPA
600/3-90/012.
Donigian, A.S., Jr. and W.C. Huber. 1991. Modeling of Nonpoint Source Water Quality in
Urban and Non-urban Areas. EPA 600/3-91/039. Environmental Research Laboratory,
Athens, GA.
Driscoll, E.D., J.L. Mancini and P.A. Mangarella. 1983a. Technical Guidance Manual for
Performing Waste Load Allocations, Book II: Streams and Rivers, Chapter 1:
Biochemical Oxygen Demand/Dissolved Oxygen. EPA 440/4-84-020.
Driscoll, E.D., T.W. Gallagher, J.L. Mancini, P.A. Mangarella, J.A. Mueller and R. Winfield.
1983b. Technical Guidance Manual for Performing Waste Load Allocations, Book II:
Streams and Rivers, Chapter 2: Nutrient/Eutrophication Impacts. EPA 440/4-84-021.
Driscoll, E.D. and D.M. DiToro. 1984. Technical Guidance Manualfor Performing Waste Load
Allocations, Book VII: Permit Averaging Periods. EPA 440/4-84-023. OWRS.
Driscoll, E.D. 1986. “Lognormality of Point and Nonpoint Source Pollutant Concentrations.”
Proceedings of Stormwater and Water Quality Model Users Group Meeting, Orlando,
FL. EPA 600/9-86/023, pp. 157-176, US EPA, Athens, GA.
Driscoll, E.D., P.E. Shelley and E. W. Strecker. 1990. SYNOP-Synoptic Rainfall Data Analysis
Program. Pollutant Loadings and Impacts from Highway Stormwater Runoff Volume I:
Design Procedure, Volume II: Users Guide, Volume III: Analytical Investigation and
Research Report, and Volume IV: Research Report Data Appendix. FHWA-RD-88-006,
007, 008, and 009 (respectively) (NTIS PB90-257551). Office of Engineering and
Highway Operations R&D, Federal Highway Administration.
Driver, N.E. and G.D. Tasker. 1988. Techniques for Estimation of Storm-Runoff Loads,
Volumes, and Selected Constituent Concentrations in Urban Watersheds in the United
States. Open-File Report 88-191. USGS, Denver, CO.
Eschenroeder, A. 1983. “The Role of Multimedia Fate Models in Chemical Risk Analysis.” In
Fate of Chemicals in the Environment. ACS Symposium Series 225. American
Chemical Society, Washington, DC.
Everhart, W.H., A.W. Eipper and W.D. Youngs. 1975. Principles of Fishery Science. Cornell
University Press, Ithaca, NY.
Fischer, H.B. 1972. “Mass Transport Mechanisms in Partially Stratified Estuaries.” Journal of
Fluid Mechanics. 53: 671-687.
Fischer, H.B., E.J. List, R.C.Y. Koh, J. Imberger & N.H. Brooks. 1979. Mixing in Inland and
Coastal Waters. Academic Press, Orlando.
Flavelle, P. 1992. “A Quantitative Measure of Model Validation and its Potential Use for
Regulatory Purposes.” Advances in Water Resources. 15: 5-13.
Freedman, P.L., D.W. Dilks and B.A. Monson. 1992. Technical Guidance Manualfor
Performing Waste Load Allocations, Book III: Estuaries, Part 4: Critical Review of
Coastal Embayment and Estuarine Waste Load Allocation Monitoring.
EPA 823-R-92-005.
Freedman, P.L. and J.K. Marr. 1990. “Receiving-water Impacts.” In Control and Treatment of
Combined Sewer Overflows. Van Nostrand Reinhold, New York. pp. 79-117.
Haith, D.A. and L.L. Shoemaker. 1987. “Generalized Watershed Loading Functions for Stream
Flow Nutrients.” Water Resources Bulletin. 23(3): 471-478.
Hinson, M.O. and D. J. Basta. 1982. “Analyzing Surface Receiving Water Bodies.” In D.J.
Basta and B.T. Bower, eds. Analyzing Natural Systems, Analysis for Regional
Residuals - Environmental Quality Management. Resources for the Future, Washington,
DC. pp, 249-388.
Huber, W.C. and R.E. Dickinson. 1988. Storm Water Management Model Version 4, User’s
Manual. EPA 600/3-88/00la (NTIS PB88-236641/AS). Environmental Research
Laboratory, Athens, GA
Hydrologic Engineering Center (HEC). 1977. Storage, Treatment, Overflow, Runoff Model
“STORM”, Users Manual (Publication No. AD-A955 817). Computer Program
723-S8-L7520. Corps of Engineers, U.S. Army, Davis, CA.
Hydroqual, Inc. 1986. Technical Guidance Manual for Performing Wasteload Allocations,
Book IV: Lakes, Reservoirs and Impoundments, Chapter 3: Toxic Substances Impact.
EPA 440/4-87-002.
International Air Transport Association (IATA). 1996. Dangerous Goods Regulations, 37th
Edition. Montreal, Quebec.
Irvine, K.N., B.G. Loganathan, E.J. Pratt and H.C. Sikka. 1993. “Calibration of PCSWMM to
estimate metals, PCBs and HCB in CSOs from an industrial sewershed.” In W. James,
ed. New Techniques for Modelling the Management of Stormwater Quality Impacts.
Lewis Publishers, Boca Raton, FL. pp. 215-242.
Jewell, T.K., T.J. Nunno and D.D. Adrian. 1978. “Methodology for Calibrating Stormwater
Models.” Journal of the Environmental Engineering Division, ASCE. 104: 485-501.
Jirka, G.H. 1992. Technical Guidance Manual for Performing Waste Load Allocations, Book
III: Estuaries, Part 3: Use of Mixing Zone Models in Estuarine Waste Load Allocations.
EPA 823-R-92-004.
Johanson, R.C., J.C. Imhoff, J.L. Kittle, Jr. and A.S. Donigian. 1984. Hydrological Simulation
Program - FORTRAN (HSPF), Users Manual for Release 8.0. EPA 600/3-84-066.
Environmental Research Laboratory, Athens, GA
Klemm. 1990. Macroinvertebrate Field and Laboratoy Methods for Evaluating the Biological
Integrity of Surface Waters. U.S. EPA. Office of Research and Development.
EPA 600/4-90/030.
LimnoTech, Inc. 1992. Simplified Method Program - Variable Complexity Stream Toxics
Model (SMPTOX3): Version 2.0 User‘s Manual.
LimnoTech, Inc. 1985. Dynamic Toxics Waste Load Allocation Model (DYNTOX: User’s
Manual.
Mancini, J.L., G.G. Kaufman, P.A. Mangarella and E.D. Driscoll. 1983. Technical Guidance
Manual for Performing Waste Load Allocations, Book IV: Lakes, Reservoirs and
Impoundments, Chapter 2, Nutrient/Eutrophication Impacts. EPA 440/4-84-019. Office
of Water Regulations and Standards, Monitoring and Data Support Division.
Mao, K. 1992. “How to Select a Computer Model for Storm Water Management.” Pollution
Engineering. Oct. 1, 1992: 60-63.
Martin, J.L., R.B. Ambrose, Jr., S.C. McCutcheon (editors). 1990. Technical Guidance Manual
for Performing Waste Load Allocations, Book III: Estuaries, Part 2, Application of
Estuarine Wasteload Allocation Models. EPA 823/R-92-003. U.S. EPA, Washington,
DC.
Mays, L.W. (editor). 1996. Water Resources Handbook. McGraw-Hill, New York, N.Y.
McKeon, T.J. and J.J. Segna. 1987. Selection Criteria for Mathematical Models Used in
Exposure Assessments: Surface Water Models. EPA 600/8-87/042. Exposure
Assessment Group, Office of Health and Environmental Assessment, U.S. EPA, July
1987.
Merritt, R.W. and K.W. Cummins. 1984. An Introduction to the Aquatic Insects of North
America, 2nd Edition. Kendall/Hunt Publishing Company, Dubuque, IA.
Metcalf & Eddy, Inc. 1991. Wastewater Engineering: Treatment, Disposal, Reuse: 3rd Edition.
McGraw-Hill, Inc., New York.
Miller, R.W. 1983. The Flow Measurement Engineering Handbook. McGraw-Hill, New York.
Mills, W.B., G.L. Bowie, T.M. Grieb, K.M. Johnson, and R.C. Whittemore. 1986.
Handbook-Stream Sampling for Waste Load Allocation Applications. EPA 625/
6-86-013. Office of Research and Development, U.S. EPA, Washington, DC.
Mills, W.B., D.B. Porcella, M.J. Ungs, S.A. Gherini, K.V. Summers, L. Mok, G.L. Rupp,
G.L. Bowie, and D.A. Haith. 1985a. Water Quality Assessment: A Screening
Procedure for Toxic and Conventional Pollutants in Surface and Ground Water -
Part I. (Revised 1985). EPA 600/6-85-002a. Environmental Research
Laboratory, U.S. EPA, Athens, GA.
Mills, W.B., D.B. Porcella, M.J. Ungs, S.A. Gherini, K.V. Summers, L. Mok, G.L. Rupp,
G.L. Bowie, and D.A. Haith. 1985b. Water Quality Assessment: A Screening
Procedure for Toxic and Conventional Pollutants in Surface and Ground Water -
Part II. (Revised 1985). EPA 600/6-85-002b. Environmental Research
Laboratory, U.S. EPA, Athens, GA.
National Council for Air and Stream Improvement (NCASI). 1982. A Study of the Selection,
Calibration and Verification of Mathematical Water Quality Models. Technical Bulletin
No. 367. New York.
Nielsen, L.A. and D.L. Johnson (editors). 1983. Fisheries Techniques. American Fisheries
Society, Bethesda, MD.
Nix, S.J. 1990. “Mathematical Modeling of the Combined Sewer System.” In Control and
Treatment of Combined Sewer Overflow. Van Nostrand Reinhold, New York, NY.
pp. 23-78.
Nix, S.J., P.E. Moffa and D.P. Davis. 1991. “The Practice of Combined Sewer Modeling.”
Water Resources Bulletin. 27(2): 189-200.
O’Connell, R.L. and N.A. Thomas. 1965. “Effect of Benthic Algae on Stream Dissolved
Oxygen.” Journal of the Sanitary Engineering Division, ASCE. 91(SA3): 1-16.
Ohio River Valley Water Sanitation Commission (ORSANCO). 1998. Fax Memorandum on
wet weather monitoring from Jim Gibson, ORSANCO, to Tim Dwyer, USEPA. Dated
March 4, 1998.
Onishi, Y. and S.E. Wise. 1982. User’s Manualfor the Instream Sediment-Contaminant
Transport Model SERATRA. EPA-600/3-82-055.
Ott, W.R. 1995. Environmental Statistics and Data Analysis. Lewis Publishers, Boca Raton,
FL.
Palmstrom, M. and W.W. Walker Jr. 1990. P8 Urban Catchment Model: User’s Guide,
Program Documentation, and Evaluation of Existing Models, Design Concepts, and
Hunt-Potowomut Data Inventory. Report No. NBP-90-50. The Narragansett Bay
Project, Providence, RI.
Pennak, R.W. 1989. Freshwater Invertebrates of the United States, 3rd Edition. John Wiley
and Sons, Inc., New York, NY.
Pitt, R. 1986. “Runoff Controls in Wisconsin’s Priority Watersheds.” In: Urban Runoff Quality
- Impact and Quality Enhancement Technology, Proceedings of an Engineering
Foundation Conference. Henniker, NH, June 23-27, 1986. ASCE, New York, NY.
pp. 290-313.
Plafkin, J.L., M.T. Barbour, K.D. Porter, S.K. Gross, and R.M. Hughes. 1989. Rapid
Rioassessment Protocols for Use in Streams and Rivers - Benthic Macroinvertebrates
and Fish. Office of Water, U.S. EPA, Washington, DC. EPA 440/4-89/001.
Reckhow, K.H. and S.C. Chapra. 1983a. “Confirmation of Water Quality Models.” Ecological
Modelling. 20: 113-133.
Reckhow, K.H. and S.C. Chapra. 1983b. Engineering Approaches for Lake Management,
Vol. 1 and Vol. 2. Butterworth Publishers, Woburn, MA.
Reckhow, K.H., J.T. Clements and R.C. Dodd. 1990. “Statistical Evaluation of Mechanistic
Water-quality Models.” Journal of Environmental Engineering, ASCE. 116(2):
250-268.
Richardson, W.L. et al. 1983. User’s Manual for the Transport and Fate Model MICHRIV.
U.S. EPA Large Lakes Research Station, Grosse Isle, MI.
Roesner, L.A., J.A. Aldrich and R.E. Dickinson. 1988. Storm Water Management Model User’s
Manual, Version 4, EXTRAN Addendum. EPA 600/3-88/001b (NTIS PB84-198431).
Environmental Research Laboratory, Athens, GA
Saunders, J.F. III, W.M. Lewis Jr. and A. Sjodin. 1993. Ammonia Toxicity Model AMMTOX
Version 1.0, Operator’s Manual. Center for Limnology, University of Colorado,
Boulder. CO
Schnoor, J.L. 1985. “Modeling Chemical Transport in Lakes, Rivers, and Estuarine Systems.”
In Neely, W.B. and G.E. Blau, eds., Environmental Exposure from Chemicals, Vol. 2.
CRC Press, Boca Raton, FL.
Schreck, C.B. and P.B. Moyle (editors). 1990. Methods for Fish Biology. American Fisheries
Society, Bethesda, MD.
Schueler, T.R. 1987. Controlling Urban Runoff A Practical Manualfor Planning and
Designing Urban BMPs. Document No. 87703. Metropolitan Washington Council of
Governments, Washington, DC.
Schwab, G.O., R.K. Frevert, T.W. Edminster, and K.K. Barnes. 1981. Soil and Water
Conservation Engineering, Third Edition. John Wiley and Sons, Inc., New York,
NY.
Shoemaker, L.L. et al. 1992. Compendium of Watershed-Scale Models for TMDL Development.
EPA 841-R-92-002. U.S. EPA Office of Wetlands, Oceans and Watersheds and Office of
Science and Technology.
Sormen, M.B. 1980. “Urban Runoff Quality: Information Needs.” Journal of the Technical
Councils, ASCE. 106(TC1): 29-40.
Terstriep, M.L., M.T. Lee, E.P. Mills, A.V. Greene and M.R. Rahman. 1990. Simulation of
Urban Runoff and Pollutant Loading from the Greater Lake Calumet Area. Prepared by
the Illinois State Water Survey for the U.S. Environmental Protection Agency, Region V,
Water Division, Watershed Management Unit, Chicago, IL.
Thomann, R.V., and J. A. Mueller. 1987. Principles of Surface Water Quality Modeling and
Control. Harper and Row Publishers, New York, NY.
U.S. Department of Agriculture (USDA) and U.S. Environmental Protection Agency (EPA).
1998. Clean Water Action Plan: Restoring and Protecting America‘s Waters.
nd
U.S. Department of the Interior (USDI), 1984. Water Measurement Manual, 2 Edition. Bureau
of Reclamation. Denver, CO.
U.S. Environmental Protection Agency (EPA). 1997. Combined Sewer Overflows-Guidance for
Financial Capability Assessment and Schedule Development. EPA 832-B-97-004.
U.S. Environmental Protection Agency (EPA). 1996. Combined Sewer Overflows and the
Multimetric Evaluation of Their Biological Effects: Case Studies in Ohio and New York.
EPA 823-R-96-002.
U.S. Environmental Protection Agency (EPA). 1995f. Guidelines for the Preparation of the
1996 State Water Quality Assessments (305(b) Reports). EPA 841-B-95-001.
U.S. Environmental Protection Agency (EPA). 1995g. Technical Guidance Manual for
Developing Total Maximum Daily Loads, Book II: Streams and Rivers, Part 1:
Biochemical Oxygen Demand/Dissolved Oxygen and Nutrients/Eutrophication.
EPA 823-B-95-007.
U.S. Environmental Protection Agency (EPA). 1994. Water Quality Standards Handbook,
Second Edition. EPA 823-B-94-006.
U.S. Environmental Protection Agency (EPA). 1994a. Combined Sewer Overflow Control
Policy.
U.S. Environmental Protection Agency (EPA). 1994c. Industrial User Inspection and Sampling
Manual for POTWs. EPA 831-B-94-001.
U.S. Environmental Protection Agency (EPA). 1994d. EPA Requirements for Quality
Assurance Project Plans for Environmental Data Operations. EPA QA-R-5.
U.S. Environmental Protection Agency (EPA). 1993. Combined Sewer Overflow Control
Manual. EPA 625-R-93-007.
U.S. Environmental Protection Agency (EPA). 1992. NPDES Storm Water Sampling Guidance
Document. EPA 833-B-92-001. Office of Water, U.S. EPA, Washington, DC.
U.S. Environmental Protection Agency (EPA). 1991a. Technical Support Document for Water
Quality-based Toxics Control. EPA 505/2-90-001. OWEP/OWRS.
U.S. Environmental Protection Agency (EPA). 1991b. Guidance for Water Quality-based
Decisions: The TMDL Process. EPA 440/4-91-001. Assessment and Watershed
Protection Division, OWRS.
U.S. Environmental Protection Agency (EPA). 1987. Guidance for Sampling of and Analyzing
for Organic Contaminants in Sediments. EPA 440/4-87-010.
U.S. Environmental Protection Agency (EPA). 1985a. Guidelines for Deriving Numerical
National Water Quality Criteria for the Protection of Aquatic Organisms. NTIS
PB85-227049.
U.S. Environmental Protection Agency (EPA). 1985b. Water Quality Assessment: A Screening
Procedure for Toxic and Conventional Pollutants in Surface and Ground Water. EPA
600/6-85/002. Environmental Research Laboratory, ORD.
U.S. Environmental Protection Agency (EPA). 1984a. Technical Support Manual: Waterbody
Surveys and Assessments for Conducting Use Attainability Analyses, Volume II:
Estuarine Systems. Office of Water, U.S. EPA, Washington, DC.
U.S. Environmental Protection Agency (EPA). 1984b. Technical Support Manual: Waterbody
Surveys and Assessments for Conducting Use Attainability Analyses, Volume III: Lake
Systems. Office of Water, U.S. EPA, Washington, DC.
U.S. Environmental Protection Agency (EPA). 1983a. Results of the Nationwide Urban Runoff
Program, Volume I, Final Report. NTIS PB84-185552.
U.S. Environmental Protection Agency (EPA). 1983b. Technical Support Manual: Waterbody
Surveys and Assessments for Conducting Use Attainability Analyses, Volume I. Office of
Water, U.S. EPA, Washington, DC.
U.S. Environmental Protection Agency (EPA). 1979. Methods for the Chemical Analysis of
Water and Wastes. EPA 600/4-79-020.
Viessman, W., Jr. and M. Hammer. 1993. Water Supply and Pollution Control. Fifth Edition.
Harper Collins College Publishers.
Viessman, W., Jr., J.W. Knapp, G.L. Lewis and T.E. Harbaugh. 1977. Introduction to
Hydrology. Second Edition. Harper and Row Publishers, New York, NY.
Vollenweider, R.A. 1969. A Manual on Methods for Measuring Primary Production in Aquatic
Environments. Blackwell Scientific Publications, Oxford, England.
Walker, J.F., S.A. Pickard and W.C. Sonzogni. 1989. “Spreadsheet Watershed Modeling for
Nonpoint-source Pollution Management in a Wisconsin Basin.” Water Resources
Bulletin. 25(1): 139-147.
Ward, R.C., J.C. Loftis and G.B. McBride. 1990. Design of Water-Quality Monitoring Systems.
Van Nostrand Reinhold, New York.
Water Pollution Control Federation (WPCF). 1989. Combined Sewer Over-low Pollution
Abatement. Manual of Practice No. FD-17. Water Pollution Control Federation,
Alexandria, VA.
Weber, C.L. et al. 1989. Short-Term Methods for Estimating the Chronic Toxicity of Effluents
and Receiving Waters to Freshwater Organisms. U.S. EPA. Environmental Systems
Laboratory. Cincinnati, OH. EPA 600/4-89-001.
Welch, E.B., R.R. Homer and C.R. Patmont. 1989. “Prediction of Nuisance Periphytic
Biomass: A Management Approach.” Water Research. 23(4): 401-405.
Wetzel, R.G. and G.E. Likens. 1979. Limnological Analyses. W.B. Saunders Company,
Philadelphia, PA.
Zander, B. and J. Love. 1990. STREAMDO IV and Supplemental Ammonia Toxicity Models.
EPA Region VIII, Water Management Division, Denver, CO.
l
The Water Environment Federation’s Combined Sewer Overflow Pollution
Abatement Manual of Practice No. FD-17 (WPCF, 1989) includes discussions on
establishing planning objectives for characterizing receiving waters, their aquatic
life, and meteorologic conditions; identifying critical events; evaluating system load
characteristics; selecting analytic methods; mapping the system; developing the
sampling plan; selecting field sampling procedures; monitoring CSS and
environmental flow; and modeling.
l
Design of Water-Quality Monitoring Systems (Ward et al., 1990) includes insightful
discussions on the design of monitoring plans, the essential role of statistics,
frameworks for designing water-quality information systems, quantification of
information, data analysis, and the documentation of monitoring plans. This
reference also includes four case studies of large-scale and long-term monitoring
programs.
l
NPDES Storm Water Sampling Guidance Document, EPA 833-B-92-001, (EPA,
1992) details EPA’s requirements for monitoring storm water discharges. When
such monitoring is required as a condition of a CSS’s NPDES permit, monitoring
efforts for CSO control should be coordinated with this required monitoring effort in
order to maximize data collection efficiencies and minimize monitoring costs.
l
A Statistical Method for Assessment of Urban Stormwater Loads, Impacts, and
Controls, EPA 440/3-79-023, (Driscoll et al., 1979) discusses approaches for
defining the purpose of monitoring programs; monitoring rainfall; using rainfall data
to project and evaluate impacts; selecting monitoring sites; characterizing drainage
basins; determining study periods, sampling frequencies, and sampling intervals
during storms; selecting sampling procedures and sampling parameters;
understanding special considerations for monitoring receiving waters; and using
continuous monitoring. It also provides an extensive literature compilation regarding
storm water and CSO monitoring.
l
Data Collection and Instrumentation in Urban Stormwater Hydrology (Jennings,
1982) reviews data and instrumentation needs for urban storm water hydrology. This
reference considers monitoring strategy design and the collection and use of data to
characterize rainfall, other meteorological characteristics, streamflows, receiving
water biologies and chemistries, and land use.
l
Use of Field Data in Urban Drainage Planning (Geiger, 1986) describes rainfall-
runoff processes and data collection constraints, the need to match data collection to
study objectives, the use of data in urban drainage planning, the application and
verification of models used in urban drainage planning, the validity of the design
storm concept, the reliability of storm water simulations, and the real-time use of
monitoring data in control and sewer system operation.
l
“Water Body Survey and Assessment Guidance For Conducting Use Attainability
Analyses (UAA).” In Water Quality Standards Handbook (EPA, 1994). The UAA
concepts discussed in this Handbook include useful field sampling methods,
modeling, and interpretation approaches in three Technical Support Documents for
flowing waters, estuaries, and lakes (EPA, 1983b, 1984a, and 1984b).
l
Several guidance documents that discuss or pertain to EPA’s Waste Load Allocation
(WLA) process also provide useful information on a wide range of topics that are
potentially valuable when planning monitoring programs for CSO control:
- Guidance for State Water Monitoring and Waste Load Allocation Programs
(EPA, 1985) includes a chapter on monitoring for water-quality-based controls.
It discusses the process of collecting and analyzing effluent and ambient
monitoring data in establishing water quality standards and EPA’s
responsibilities in this process.
Table A-l
Checklist of Considerations for Documenting Monitoring
Program Designs and Implementation (expanded from Ward et al., 1990)
Pre-Sampling Preparations
Selecting personnel and identifying responsibilities
Training personnel in safety and confined space entry; verifying first aid and wet-weather training,
CPR, currency of vaccinations etc.)
Preparing site access and obtaining legal consents
Acquiring necessary scientific sampling or collecting permits
Developing formats for field sampling logs and diaries
Training personnel in pre-sampling procedures (e.g., purging sample lines, instrument calibration)
Checking equipment availability, acquisition, and maintenance
Scheduling sample collection (random? regular? same-time-of-day?)
Preparing pre-sampling checklist
Sampling Procedures
Procedures documentation
Staff qualifications and training
Sampling protocols
Quality-control procedures (equipment checks, replicates, splits, etc.)
Required sample containers
Sample numbers and labeling
Sample preservation (e.g, “on ice” or chemical preservative)
Sample transport (delivery to laboratory)
Sample storage requirements
Sample tracking and chain-of-custody procedures
Quality control or quality assurance
Field measurements
Field log and diary entries
Sample custody and audit records
Post-Sampling Follow Up
Filing sample logs and diaries
Cleaning and maintaining equipment
Disposing of chemical wastes properly
Reviewing documentation and audit reports
Laboratory Analysis
Sample Analysis
Sample analysis methods and protocols
Use of reference samples, duplicates, blanks, etc.
Quality control and quality assurance compliance
Sample archiving
Proper disposal of chemical wastes
Full documentation in bench sheets
Data Management
Data Analysis
Selecting software
Handling missing data and non-detects
Identifying and using data outliers
Planning graphical procedures (e.g., scatter plots, notched-box and whisker)
Parametric statistical procedures
Non-parametric statistical procedures
Trend analysis procedures
Multivariate procedures
Quality control checks on statistical analyses
Reporting
Information Use
General
Contingencies
Follow-up procedures
Data management
Data analysis
Reporting
Information use
Table A-2
Checklist for Reviewing CSO Monitoring Plans
Up-to-date
Shows “as-built” sewer system
Shows drainage areas with land use information
Shows location of major industrial sewer users
Shows location of all direct discharge points, including all related CSO, POTW, storm water, and
industrial discharges
Distinguishes bypass points from CSOs points and shows locations
Shows locations of CSO quantity and quality monitoring sites
Identifies receiving waters
Identifies designated and existing uses of receiving waters
Shows areas of historical use impairment
CSO Volume
CSO Quality
Table B-l
Documents and Screening Manual (Mills et al.) for Analysis of Conventional Pollutants
Algal
Predictions Algal
Streeter- Without Predictions Algal Effects Algal Effects
Phelps DO NH3 Toxicity Nutrient With Nutrient on Daily on Diurnal
Data Requirements Analyses-a Calculations-b Limitations-c Limitations-c Average DO-c DO-c
Hydraulic and Geometric Data
Flow Rates-d x x x x x x
Velocity x x x x x x
Depth x x x x x x
Cross-sectional area x x x x x x
Reach length x x x x x x
DO x
CBOD, NBOD x
NH3 x
Temperature x x x x x x
Inorganic P x x x x
Inorganic NPDES x x x x
Chlorophyll a f x x x x
pH x
DO/BOD Parameters
January 1999
Appendix B
Algal Algal
Predictions Predictions Algal Effects
Streeter- Without With on Daily Algal Effects
Phelps DO NH3 Toxicity Nutrient Nutrient Average on Diurnal
Data Requirements Analyses-a Calculations-b Limitations-c Limitations-c DO- c DO- c
Phytoplankton Parameters
Maximum growth rate x X X X
Respiration rate X X X X
Settling velocity X x x X
Phosphorous half- X X X
saturation constant
Nitrogen half-saturation X X X
Phosphorous to X X X X
chlorophyll ratio
X X X X
ratio
Light Parameters
X X X X
Light extinction X X X X
coefficient
a)
Streeter-Phelps DO calculations are described in Chapter 1 of Book II of the WLA guidance documents
(Table 1- 1) and the Screening Manual (Mills et. al.).
b)
Ammonia toxicity calculations are described in Chapter 1 of Book II of the WLA guidance documents.
c)
Algal predictions and their effects on DO are discussed in Chapter 2 of Book II of the WLA guidance documents.
d)
Flow rates are needed for the river and all point sources at various points to define nonpoint flow,
e)
Constituent concentrations are needed at the upstream boundary and all point sources.
f)
Chlorophyll a concentrations are also needed at the downstream end of the reach to estimate net growth rates,
Table B-2
Model Input Parameters for Qual-2E
January 1999
Appendix B
Water Quality
Water Variable Spatial Parameters
Model Quality Hydraulics Loading Rated Types of Loads Dimensions Water Body Modeled Chemical/Biological Physical
DOSAG-I Steady- Steady-state No multiple point I-D stream network DO, CBOD, 1st-order decay of dilution,
state source NBOD, NBOD, CBOD, advection,
conservative coupled DO reservation
SNSIM Steady- Steady-state No multiple point I-D stream network DO, CBOD, 1st-order decay of dilution,
state sources & NBOD, NBOD, CBOD, advection,
nonpoint sources conservative coupled DO, benthic reservation
demand (s),
photosynthesis (s)
QUAL-II Steady- Steady-state No multiple point I-D stream network DO, CBOD, 1st-order decay of dilution
state or sources & temperature, NBOD, CBOD, advection,
dynamic nonpoint sources ammonia, nitrate, coupled DO, benthic reservation, heat
nitrite, algae, demand (s), CBOD balance
phosphate, settling (s), nutrient-
coliforms, non- algal cycle
conservative
substances, three
conservative
substances
RECEIV-II Dynamic Dynamic Yes multiple point 1-D or 2-D stream network DO, CBOD, 1st-order decay of dilution,
sources or well-mixed ammonia, nitrate, NBOD, CBOD, advection,
estuary nitrite, total coupled DO, benthic reservation
nitrogen, demand (s), CBOD
phosphate, settling (s), nutrient-
coliforms, algae, algal cycle
salinity, one
metal ion
(s) = specified.
Appendix B
Table B-4
Methods for Determining Coefficient Values in Dissolved Oxygen
and Eutrophication Models
O2 production per unit photosynthesis Literature values, model calibration and measurement
by light to dark bottles and chambers.
Ammonia Parameters
Nitrite Parameters
Nitrite oxidation rate Use literature values and calibration, since this rate is
much faster than the ammonia oxidation rate.
Phosphate Parameters
Phosphorous fraction of algal biomass Literature values and model calibration or laboratory
determinations from field samples.
Table B-5
Summary of Data Requirements for Screening Approach for Metals in Rivers
Calculation
Data Methodology Remarks
Hydraulic Data
1. Rivers:
l
River flow rate, Q D, R, S, L An accurate estimation of flow rate is very important because of
dilution considerations. Measure or obtain from USGS gage.
l Cross-sectional area, A D, R, S
l Water depth, h D, R, S, L Average water depth is cross-sectional area divided by surface
width.
l Reach lengths, x R, S
l Stream velocity, U R, S Required velocity is distance divided by travel time. It can be
approximated by Q/A only when A is representative of the reach
being studied.
2. Lakes:
l Hydraulic residence time, L Hydraulic residence times of lakes can vary seasonally as the flow
T rates through the lakes change.
l
Mean depth, H L Lake residence times and depths are used to predict settling of
absorbed metals in lakes.
Source data
1. Background
l
Metal concentrations, Ct D, R, S, L Background concentrations should generally not be set to zero
without justification.
2. Point sources
l
Locations D, R, S, L
l
Flow rate, Qw D, R, S, L
l
Metal concentration, Ctw D, R, S, L
l Suspended solids, SW D, R, S, L
Bed Data
l
Depth of contamination For the screening analysis, the depth of contamination is most
useful during a period of prolonged scour when metal is being
input into the water column from the bed.
l
Porosity of sediments, n
l
Density of solids in sediments (e.g., 2.7 for sand) us
l Metal concentration in bed during prolonged scour period, Ct2
Derived Parameters
l
Partition coefficient, Kp All Partition coefficient is a very important parameter. Site-specific
determination is preferable.
l Settling velocity, ws S,L Parameter derived based on suspended solids vs. distance profile.
l Resuspension velocity, W rs R Parameter derived based on suspended solids vs. distance profile.
Equilibrium Modeling
l Water quality E Equilibrium modeling is required only if predominant metal
characterization of river: species and estimated solubility controls are needed.
l pH
l Suspended solids
l Conductivity
l Temperature
l
Hardness Water quality criteria for many metals are keyed to hardness, and
allowable concentrations increase with increasing hardness.
L - lake.